# VibeControls — full content > One control plane for multi-agent orchestration, secure localhost tunneling, persistent tmux sessions, and browser terminals. RBAC, BYOK on AI, 50+ plugins. Generated from the site's own pages at build time. 144 pages. --- # VibeControls — Agentic Engineering Mission Control > One control plane for multi-agent orchestration, secure localhost tunneling, persistent tmux sessions, and browser terminals. RBAC, BYOK on AI, 50+ plugins. Source: https://vibecontrols.com/ ## Mission Control for AI-Native engineering One console for every machine, agent, terminal, tunnel, and AI assistant your team runs. Spin up reproducible vibes , share live sessions, and let agents handle the boring parts — local laptop to global production. ### See VibeControls in action A demo and deep-dive on the agent + plugin model and the public roadmap — straight from the team. ### Built for the work VibeControls customers actually do Real scenarios, start to finish — pick one to see how it plays out. #### Zero-config local agent in under a minute From nothing to a unified control plane with one command 45 scenarios across 5 areas of VibeControls. ### Bring your own coding agents. Keep control of every machine they touch. Claude Code, Codex, Gemini and more, running on your machines with your keys — in one console with per-session limits, an approval step, and a record of everything they did. #### Run Claude Code, Codex, Gemini and OpenCode from one console Every coding agent on every machine you own, in one tabbed console with per-turn model, mode and tool-reach controls. 12 shipped · 1 in progress · 1 on the roadmap ### Built for the way you actually work From a solo developer on one laptop to a platform team governing a fleet — pick your role and see what changes. #### A faster, visual replacement for tmux + SSH + a wiki Full-stack, backend, and DevOps engineers who juggle three to five environments at once. Stop wiring up brittle scripts by hand. - One dashboard for every machine you develop on - Browser-native terminals — no SSH client to babysit - Per-vibe env vars, notes, and one-click decks travel with the project ### Everything you need to control your remote environments, SSH tunnels, and terminal sessions One control plane for multi-agent orchestration, secure localhost tunneling, persistent remote tmux sessions, and browser-based terminal clients — governed by per-environment RBAC. #### Vibe Agents Lightweight agents on local, remote, container, or cloud machines with secure real-time communication. #### Secure Tunnels Port forwarding with local-to-remote, remote-to-local, and dynamic SOCKS proxy tunneling. #### Plugin Marketplace 50+ VibeModules including database GUI, container manager, code editor, and AI tools. #### Terminal Sessions tmux, SSH, and custom terminal sessions with real-time output streaming and persistence. #### Code Server Monaco-based code editor in your browser with syntax highlighting and IntelliSense. #### Sandbox Isolated development environments for safe experimentation and testing. #### Workflow Automation & Scheduling Cron-style scheduled tasks running on the agent: webhooks, scripts, AI prompts, and sprint hooks — no separate scheduler service. #### Team Collaboration Share vibes with view/edit/admin permissions, team workspaces, and RBAC. ### Problems we solve Engineering teams waste hours on environment friction. VibeControls removes it. #### Terminals, tunnels, and sessions scattered across machines One control plane for every agent, tmux session, SSH target, and cloudflared tunnel — searchable, shareable, and audited. #### Localhost services that only work on your laptop Spin up secure local→remote and remote→local tunnels in seconds, with dynamic SOCKS proxies for full network flexibility. #### Every team rebuilds the same DevOps glue An open plugin SDK with 50+ VibeModules — database GUI, container manager, code-server, GitOps, AI providers, and more. Marketplace coming soon. #### Production access without guardrails Role-based access control enforced at the API edge, detailed audit logging, workspace isolation, and human-in-the-loop approvals for sensitive actions. #### AI agents that can't touch real infrastructure MCP server, BYOK on 16+ providers, and agent-safe sandboxes so Claude, Codex, and Cursor can drive VibeControls directly. #### Onboarding engineers to complex environments Reproducible vibes, shared sessions, and live terminal sharing so new teammates ship on day one instead of day fourteen. ### Meet your environments where you already are VibeControls is API-first — every capability is reachable from every surface. Pick the client that fits the moment; the workspace, identity, and audit trail stay the same. #### Web app The full product — dashboards, agents, vibes, sessions, docs, AI, audit, webhooks, sandboxes, catalog. #### Desktop app Code-signed, notarized Electron builds for Linux, macOS, and Windows from vibecontrols.com/ecosystem/desktop. #### Mobile app Capacitor iOS + Android builds for on-the-go session monitoring and approvals. #### CLI Terminal-native automation and CI/CD integration with 175+ subcommands and an interactive TUI mode. #### Browser extension Quick stats, one-click actions, and monitoring without leaving your browser (Chrome MV3). #### VS Code extension Explorer sidebar, session lifecycle, and in-editor terminal access right where you code. #### MCP server 600+ tools, plus resources and prompts for Claude Desktop, Claude Code, and any MCP-compatible agent. #### Node + Python SDKs Typed, async-first programmatic access for automation, scripts, and serverless functions. #### Community hub Apache-Answer Q&A at community.vibecontrols.com, SSO’d against your workspace identity. ### Let agents do the boring parts. Keep humans in the loop where it matters. Autonomous agents deploy code into **isolated sandboxes**, run tests through **local tunnels**, and request **human-in-the-loop terminal approval** before promoting to alpha or prod. Every step is RBAC-checked and audited. - BYOK on AI. 16+ providers — Claude, Codex, Copilot, Gemini, Aider, OpenCode, Plandex, OpenAI-compatible endpoints. - Cited responses with file/line provenance. Plan mode, code mode, and ask mode. - MCP server so Claude Desktop, Claude Code, and any MCP-compatible agent can drive the platform directly. ### Run it yourself — or let our team help Beyond the product, we offer expert services: get more from VibeControls with setup, consulting, and training — or hand us done-for-you outcomes powered by it. #### Deployment & Setup Agents across local, remote, and cloud — wired into a clean, observable topology from day one. #### Custom Plugins New VibeModules, providers, and Vibe Deck actions built against the three-tier plugin model. #### Security & Governance Least-privilege RBAC, BYOK encryption, security plugins, and scorecard baselines. #### Training & Enablement Role-based workshops, golden-path templates, and onboarding so teams ship in hours. ### Get started in minutes Three simple steps to take control of your development environment. #### Install Agent Install the lightweight VibeControls agent on your local machine, remote server, or cloud instance in seconds. #### Connect & Configure Connect your agent to VibeControls, add your projects as vibes, and install plugins from the marketplace. #### Control Everything Manage terminals, tunnels, sessions, and team access from a single unified dashboard. ``` # 1 · install the agent npm install -g @vibecontrols/agent # 2 · start it (registers with your workspace) vibe start # 3 · keep it running across reboots vibe autostart install ``` Minimal requirements: Linux, macOS, or Windows with a supported JavaScript runtime and outbound internet. Plugin prerequisites are auto-detected. ### You install an agent on production. We treat that like adults. Conservative defaults, RBAC at the API edge, immutable 7-year audit logging, AES-256-GCM at rest, TLS in transit. SOC 2 + ISO 27001 in flight. #### AES-256-GCM at rest Agent SQLite + control-plane KMS-wrapped secrets. Credentials never travel in plaintext. #### TLS in transit Agent ↔ control plane is HTTPS + WSS. Tunnels terminate TLS at the Cloudflare edge. #### Workspace RBAC Enforced at the API edge. Identity claims are verified at the gateway so backend services never re-implement access control. #### Workspace isolation Multi-tenant isolation by workspace. Self-host on Enterprise for dedicated tenancy. #### Immutable 7-year audit log Every mutation captured with actor identity, target resource, and timestamp. Retained for 7 years with tamper-evident batch logging. SOC 2 + ISO 27001 evidence pipeline targeted. SIEM streaming planned. #### Open source agent The agent that runs on your machine is open. Read the code, audit the signed releases, pin shas. ### Pricing Plans Start free on SaaS. Dedicated tenant + self-hosting included with Enterprise. #### VibeControls - Free Choose this plan - ✓ 1 agent - ✓ 1 workspace · 5 vibes - ✓ 3 active sessions (2h limit) - ✓ 1 SSH target · 1 tunnel - ✓ 5 AI sessions/day (BYOK) - ✓ 500 MB storage - ✓ Community support #### VibeControls - Hobby (Monthly) Choose this plan - ✓ 5 agents - ✓ 1 workspace · 10 vibes - ✓ 10 active sessions (4h limit) - ✓ 7 plugin installs - ✓ 25 AI sessions/day (BYOK) - ✓ 5 GB storage - ✓ Email support #### VibeControls - Pro (Monthly) Choose this plan - ✓ 20 agents - ✓ 3 workspaces · 100 vibes - ✓ Unlimited sessions - ✓ 50 SSH targets · 25 tunnels - ✓ 100 AI sessions/day (BYOK) - ✓ 25 GB storage - ✓ Priority support (<24h) - ✓ API access ### Questions, answered The essentials before you start. The full, searchable FAQ has the rest. #### What is VibeControls? #### Who is VibeControls for? #### Do I need to use VibeControls' AI infrastructure to run prompts? #### Is VibeControls open source? #### How does VibeControls pricing work? #### How do I access VibeControls? ### Ready to take control of your development environment? Join developers who are already using VibeControls to streamline their workflow and boost productivity. Built on the Burdenoff platform — inherited in every edition - OIDC SSO - Role-based access control - Multi-tenancy - Detailed audit logging - Billing & quotas - In-app · email · webhook notifications - Internationalization - Accessibility & command palette - Dark / light / high-contrast themes - Conversations & comments - Tagging - Feature flags - Shared file storage - Developer portal - In-app store - Support & ticketing - Scheduler & workflow - Tours & onboarding - OIDC SSO - Role-based access control - Multi-tenancy - Detailed audit logging - Billing & quotas - In-app · email · webhook notifications - Internationalization - Accessibility & command palette - Dark / light / high-contrast themes - Conversations & comments - Tagging - Feature flags - Shared file storage - Developer portal - In-app store - Support & ticketing - Scheduler & workflow - Tours & onboarding --- # About - VibeControls > Discover VibeControls: Agentic Engineering Mission Control with AI-native environments, agents, sessions, and plugin runtime on Burdenoff Workspaces. Source: https://vibecontrols.com/about ## About VibeControls The Agentic Engineering Mission Control for AI-Native teams — a unified control plane for environments, terminal sessions, agents, AI tools, and operational workflows across local, remote, and cloud machines. ### What is VibeControls VibeControls is a unified **Agentic Engineering Mission Control** that orchestrates the environments, terminal sessions, agents, AI tools, and operational workflows a developer or team uses every day — across local, remote, and cloud machines — from a single workspace-native interface. Modern developers juggle three to five environments at once (local, staging, production, client servers, sandbox VMs). Traditional tools either lock you into the cloud or leave you wiring up tmux, SSH, and a pile of scripts by hand. VibeControls sits in the middle: an **agent-based architecture** that runs anywhere, a **plugin ecosystem** for everything else, and a **workspace-aware** backend that gives teams shared visibility, governance, and collaboration. #### What you get - • One dashboard for every machine you develop on - • Real-time terminal sessions you can share with teammates - • A growing catalog of plugins for terminals, tunnels, AI assistants, code editors, and integrations - • Programmable surfaces: web app, CLI, browser extension, VS Code extension, MCP server, and Node/Python SDKs - • Audit, RBAC, billing, notifications, and quotas inherited from the Burdenoff platform ### Vision A world where every developer commands their entire engineering universe from one console — local laptop to global production — and where the distance between intent and impact collapses to a single, governed click. VibeControls exists to dissolve the friction tax modern software engineering pays daily: the context-switches, the brittle scripts, the tribal knowledge, the tools that do not talk to each other, the governance bolted on after the damage is done. We see a future where every machine a developer touches is a first-class citizen — observable, governable, collaborative — and where AI is a teammate, not a tab. ### Mission To give every developer and every engineering team a unified, agent-native control plane that turns multi-environment software development from a daily struggle into a continuous flow of value — with governance, intelligence, and extensibility built in from day one. We pursue this mission by starting where developers actually work (the agent on every machine, not a catalog on a server), earning adoption through utility first and governance second, opening every surface (Web, CLI, MCP, SDKs, browser, editor), and treating extensibility as a first principle — so the ecosystem becomes the product. ### Who VibeControls is for From individual power users to enterprise platform teams. #### Individual power users Full-stack, backend, and DevOps engineers who want a faster, more visual replacement for "tmux + SSH + a wiki". #### Development teams Teams that need shared environments, onboarding consistency, and operational visibility across staging, production, and customer servers. #### DevOps & platform engineers Managing fleets of staging, production, and customer servers with port-forwarding, scheduled tasks, and runbooks. #### Freelancers & consultants Switching across many client projects and remote environments per week — without losing context. #### Educators & bootcamps Teaching modern dev workflows with shareable, reproducible environments and live session collaboration. #### Research & HPC teams Coordinating long-running sessions across shared compute infrastructure with audit and governance. ### Core concepts The vocabulary of VibeControls. Every feature is built on these primitives. #### Target A machine you want to work on — your laptop, a VM, a remote server, or a container host. Targets can be DIRECT (localhost) or SSH (remote). VibeControls models the connections between them as a graph so you always know what reaches what. #### Agent A lightweight runtime you install on a target. The agent runs locally, owns plugin execution, manages terminal sessions, and exposes a secure API. Agents register with your workspace and report health continuously. #### Vibe A project context — a workspace-scoped grouping of related work. A vibe maps to a folder on a target, can nest under a parent vibe, and carries metadata: git remote, env vars, tags, notes, attached docs, and operational tooling. #### Session A terminal or script execution running on an agent. Sessions are provider-aware (tmux, wezterm, zellij, SSH, or custom runners), persist across reconnects, stream to a browser terminal, and can be shared with teammates. #### Control A discrete operational surface — one purposeful tool that does one job well. The Dashboard, Vibe Deck, Workflow Automation, Audit log, and AI hub are all controls. The product name is plural because VibeControls is the collection of these controls — and plugins can add new ones. #### Plugin The extensibility model. Every non-trivial capability — session providers, tunnels, AI tools, editors, integrations — ships as a plugin. Plugins follow a strict three-tier model: meta interfaces, concrete providers, and standalone tools. #### Workspace The Burdenoff container that owns everything: members, roles, agents, vibes, billing, audit, notifications. VibeControls inherits multi-tenancy, RBAC, and governance from the workspace — you do not have to set any of it up yourself. ### Our core values The principles that guide everything we do at VibeControls. #### Developer first Every decision is driven by what makes developers more productive and happier in their work. The agent runs on every machine because that is where developers actually live. #### Workspace-native Every agent, vibe, session, note, and document belongs to a workspace, so collaboration, permissions, and audit are built in from day one — not bolted on later. #### Extensibility first Meta plugins, provider plugins, and an open SDK turn the ecosystem into the product. No feature gating. Regionally fair, quota-based pricing. ### Meet the team VibeControls is built by the team at Algoshred Technologies Pvt. Ltd. ### About our company VibeControls is proudly developed by Algoshred Technologies Pvt. Ltd., part of the Burdenoff Group. #### Algoshred Technologies Pvt. Ltd. **Address:** Plot No.43, Veeramani Nagar, 2nd Cross Street, Nanmangalam, Chennai - 600117, India **Website:** [https://vibecontrols.com](https://vibecontrols.com) **Contact:** [Get in touch](https://vibecontrols.com/contact) ##### Our commitment We are committed to empowering developers worldwide with innovative tools that simplify complex workflows. Our mission is to make development-environment management effortless and efficient for teams of all sizes. ### Ready to contribute to the future? Join our community of developers who are building the next generation of development tools. --- # AI Native — VibeControls > Coding agents on your own machines with your own keys, in one console: per-session limits, a machine that starts locked, a plan you approve, a live record of every AI action, and honest notes on what is still on the roadmap. Source: https://vibecontrols.com/ai ## Bring your own coding agents. Keep control of every machine they touch. Claude Code, Codex, Gemini and more, running on your machines with your keys — in one console with per-session limits, an approval step, and a record of everything they did. A coding agent needs a machine to run on, a model to call, a project to understand and a line it cannot cross — here, all four are yours to set. Anything you can do, an agent can do through the same controls: never more, and never without leaving a record. Everything below is labelled with where it actually stands in Vibecontrols today — shipped, in progress, or on the roadmap. ### AI-native by architecture, not bolted on Four architectural choices that make AI a first-class part of Vibecontrols rather than a chat window on the side. #### One set of controls for you and your agents Anything an agent does goes through the same permission check you do, so it can never reach further than your own access — and the record names you. #### Your models, on your machines Coding agents run on machines you own and read your key from that machine — eight providers, including OpenAI-compatible endpoints and local Ollama. #### Autonomy is a dial, not a switch Choose per session: plan only, apply edits, or full auto. Every machine also refuses remote commands until you unlock that machine yourself. #### Every AI action leaves a record A live feed of what your coding agents did, kept apart from the log of who changed what, plus a per-message cost you see before you send. ### From intent to a governed action Walk the path a request takes. Select any stage to see what happens there and what backs it. Stage 1 of 5 · Context #### The page you're on is part of the question The thing you're looking at, plus contexts, prompts and instruction files you choose. Open the assistant anywhere and it already knows which agent, vibe or session you mean — no ids to paste. Beyond that you pick the grounding: saved contexts, shared prompt templates, and instruction files every agent on the project reads. Nothing is guessed. ### What the AI in Vibecontrols actually does Filter by delivery status, then open any capability for the detail and what backs it. No capability is listed as shipped without something in the product behind it. ### Where the AI shows up The places AI meets the work in Vibecontrols — and how far each one has actually got. #### AI workbench The in-app console where your coding-agent sessions live: three panes, tabs across machines, and every turn configurable. - Tabbed live sessions across several machines, each with its own link - Per-turn model, mode, MCP reach, working folder, file attach and voice - Admin overlay for templates, contexts, prompts, tasks, stats and keys ### See it as a real scenario Every AI capability above shows up in a concrete Vibecontrols story. Open one to read the full walk-through. ### AI you can actually let near your data An AI-native product has to be governable. Here is where Vibecontrols stands on each control — including the parts still being built. #### The assistant cannot exceed you Every action is checked against your own permissions and runs under your login — so what you cannot do, it cannot do, and the record names you, not a bot. #### Every machine starts locked A machine refuses commands, scripts and file changes that did not start locally until you unlock it — and commands and file edits unlock separately. In doubt, it refuses. #### Full autonomy takes an explicit yes Full auto cannot be chosen quietly: it asks first, and is not offered for chat-only sessions with nothing to grant. Where plan review is set up, the plan comes first. #### Two trails, kept apart One feed records what the AI coding tools did on your machines; the audit log records who changed what. Keeping them apart means neither question buries the other. #### Context boundary audit Only references leave with your question today. Still in progress: the written proof that vibe secrets, raw terminal output and access tokens never travel with a prompt. #### Test set and release gate No test set of common questions exists yet, and no gate running one before a prompt or model change, including probes for instructions hidden in notes. Planned only. #### Models Two paths, kept visibly apart. Sessions on your own machines call the model with your key, across eight providers plus any OpenAI-compatible endpoint, and are not charged here. The built-in assistant runs on workspace compute and is charged in AI credits. - Anthropic Claude - OpenAI / Codex - Google Gemini - OpenCode - Cursor - OpenRouter - Minimax - Ollama (cloud or self-hosted) - Any OpenAI-compatible endpoint (Together, Groq, Mistral, self-hosted) ### The honest answers The split is deliberate. A session on your own machine sends the prompt to that machine, which calls the provider with a key kept there — we never need a copy. The built-in assistant is different: it runs on workspace compute, which is why it is charged in AI credits and your own sessions are not. ### Put Vibecontrols’s AI to work Claude Code, Codex, Gemini and more, running on your machines with your keys — in one console with per-session limits, an approval step, and a record of everything they did. --- # Engineering Blog - VibeControls > Technical insights, product updates, and engineering deep dives from the VibeControls team. Source: https://vibecontrols.com/blog ## Engineering Blog Technical insights, product updates, and behind-the-scenes from the VibeControls team #### How We Built VibeControls: Architecture Deep Dive A look under the hood at the microservices architecture, GraphQL Federation, and infrastructure powering VibeControls. #### Introducing VibeControls — Mission Control for Developers We built VibeControls to solve a problem every developer faces: managing the chaos of modern development environments. #### 50+ Plugins: The VibeControls Ecosystem Explore the VibeControls plugin marketplace with 50+ extensions that supercharge your developer workflow. #### Choosing the Right Developer Plugins for VibeControls VibeControls' plugin system spans AI orchestration, security, GitOps, sessions, storage, and tunnels — here's how the tiers work and how to pick the right combination for your team. --- # Architecture Deep Dive — VibeControls > Technical deep dive into the microservices architecture behind VibeControls. Source: https://vibecontrols.com/blog/architecture-deep-dive ## How We Built VibeControls: Architecture Deep Dive VibeControls is built on modern microservices designed for real-time collaboration. ### The Stack - Bun + TypeScript - ElysiaJS + GraphQL Yoga - PostgreSQL + Prisma - Valkey (Redis fork) - Kubernetes + Istio - React 19 + Vite + TailwindCSS v4 ### Agent Architecture Lightweight agents connecting via WebSocket for real-time bidirectional communication. ### Plugin System Self-contained modules with manifest, backend handler, and UI component. ### Security mTLS via Istio, RBAC, audit logging, AES-256 credential encryption at every layer. --- # Introducing VibeControls > We built VibeControls to solve a problem every developer faces: managing multiple projects, terminals, tunnels, and environments from one place. Source: https://vibecontrols.com/blog/introducing-vibecontrols ## Introducing VibeControls — Mission Control for Developers Every developer juggles multiple tools — terminals, SSH clients, tunnel managers, code editors. VibeControls is the developers command center. One dashboard to manage everything. ### What VibeControls Does #### Vibe Agents Lightweight agents on local, remote, Docker, or cloud machines with real-time WebSocket control. #### Terminal Sessions tmux, SSH, and custom terminals with real-time streaming and persistence. #### Secure Tunnels Port forwarding with local-to-remote, reverse, and SOCKS proxy support. #### 50+ Plugins PostgreSQL GUI, Docker Manager, Code Editor, AI tools, and more. #### Team Collaboration Share vibes, RBAC, shared sessions, team workspaces, and audit logging. ### Get Started [Check pricing](https://vibecontrols.com/pricing) — Free, Hobby, and Pro plans available now. --- # Plugin Ecosystem — VibeControls > From database GUIs to AI assistants — explore the VibeControls plugin marketplace. Source: https://vibecontrols.com/blog/plugin-ecosystem ## 50+ Plugins: The VibeControls Ecosystem VibeControls ships with 50+ plugins and a Plugin SDK to build your own. ### Database Tools - PostgreSQL GUI - Redis/Valkey Inspector - MongoDB Explorer ### Container Management - Docker Manager - Kubernetes Dashboard ### Development - Code Server (Monaco editor) - Git Integration - API Tester ### AI and Automation - Claude AI Assistant - Workflow Automation ### Build Your Own The Plugin SDK provides storage APIs, networking, UI components, and lifecycle hooks. --- # Choosing the Right VibeControls Plugins > VibeControls ships 50+ open-source plugins for AI agents, security, GitOps, sessions, and tunnels. Here Source: https://vibecontrols.com/blog/vibecontrols-plugins-review ## Choosing the Right Developer Plugins for VibeControls VibeControls plugins turn one lightweight agent into exactly the toolkit your team needs, and nothing you don't. Instead of shipping a bloated all-in-one binary, VibeControls keeps its core agent small and lets you install only the AI providers, security checks, terminals, storage backends, and network tools your workflow actually calls for. This guide walks through every plugin category that ships today, explains what each one really does, and gives you a practical way to decide which ones belong on your machines. ### What Are VibeControls Plugins? A VibeControls plugin is a small, installable module that extends the `vibe` agent — the background process that runs on your laptop, a remote server, a container, or a cloud VM. Instead of bundling every possible integration into the core binary, the agent ships thin and you add capability with a single command: ``` vibe plugin install @vibecontrols/vibe-plugin-ai ``` The plugin ecosystem is over 50 packages and growing, and every one of them is MIT licensed and published on GitHub — you can read the source, fork it, or ship your own alongside it. The agent runtime that loads the plugins is closed source, but the plugins themselves are not. That split matters: it means the parts of the system most likely to need a fix, an update, or a community contribution are the parts anyone can inspect. **In short:** the best VibeControls plugins are the ones that match your stack. AI orchestration plugins if you switch between coding agents, security plugins if you need lifecycle-stage checks, session plugins if you live in a terminal, and storage or tunnel plugins depending on how you share state and expose services. ### The Three Plugin Tiers, Explained Plugins aren't a flat list — they're organized into three tiers, and understanding the difference makes the catalogue much easier to navigate. | Tier | What it does | Example | | --- | --- | --- | | **Meta** | Defines one interface, then routes to whichever provider you install | `vibe-plugin-ai`, `vibe-plugin-security` | | **Provider** | Implements a meta plugin's interface for one specific backend | `vibe-plugin-ai-claude`, `vibe-plugin-storage-postgres` | | **Tool** | Stands alone — no meta interface needed | `vibe-plugin-tool-ssh`, `vibe-plugin-tool-code-server` | There are seven meta plugins today: AI, Security, GitOps, Plan, Storage, Session Manager, and Tunnel. Each one owns a single contract so that switching providers — say, moving from tmux to Zellij, or from Cloudflare Tunnel to a self-hosted relay — never touches the rest of your setup. You install a different provider, and the meta plugin routes to it. ### AI Provider Plugins: Run Any Coding Agent You Want The AI meta plugin is the one most developers install first. It gives VibeControls one consistent interface for talking to AI coding agents, and you decide which providers sit behind it. At last count, more than a dozen provider plugins are published, covering names most developers already recognize: Claude (Anthropic), Codex (OpenAI), Gemini (Google), GitHub Copilot, Cursor, Aider, Sourcegraph Amp, Block Goose, OpenRouter, Ollama, and any OpenAI-compatible endpoint for self-hosted or proxied models. Each one is bring-your-own-key (BYOK), so your API credentials and your prompts stay under your control rather than routed through a third party. Why does this matter more than it sounds? Teams rarely settle on a single AI provider forever. Pricing changes, a new model ships, or a client mandates a specific vendor. Because the AI plugin is a meta interface, you can install a second or third provider and pick per session which one drives a given task — without rewriting how your agent, your scripts, or your team's habits work. ### Session and Terminal Plugins for a Browser-Native Workflow VibeControls' interface layer is built around one idea: you should be able to drive a real terminal, a real editor, and a real Git client from a browser tab, with the session staying alive even if you close the laptop. The Session Manager meta plugin handles capability discovery and provider routing across three multiplexers: - **tmux** — the default, paired with `ttyd` for persistent, shareable terminal sessions - **WezTerm** — a GPU-accelerated multiplexer for heavier terminal workloads - **Zellij** — a terminal workspace built around floating panes On top of sessions, a set of standalone tool plugins add specific browser-based capabilities without needing a meta interface at all: - **code-server** — a full VS Code instance in the browser, reverse-proxied through the agent - **Git client (Ungit)** — a visual Git interface for teams that want to see history and diffs without a local GUI app - **GraphiQL** — an in-browser API playground for exploring any GraphQL endpoint - **SSH** — connection management, remote terminals, port forwarding, and remote agent installation, all from the dashboard Together, these plugins are what let a session started on a work laptop get picked back up, unchanged, from a phone browser or a teammate's machine — the session lives on the agent, not the client. ### Storage Plugins: Choosing Where Agent State Lives Every agent needs somewhere to keep its state — session history, plugin configuration, credentials. The Storage meta plugin owns a single `AgentDatabase` contract, and two providers implement it today: - **Skalex** — an embedded, document-oriented, file-backed store. Good for a single machine that doesn't need to share state with anything else. - **Postgres** — a shared backend with full feature parity to Skalex, used when multiple agents need to read and write the same state (for example, a team sharing session history across several remote machines). Both providers encrypt data at rest with AES-256-GCM, so the choice between them is really about topology, not security: one machine, one agent, embedded storage is simplest; several agents sharing state, Postgres is the provider built for that. ### Security and Tunnel Plugins: Governed Remote Access Two categories exist specifically because remote development multiplies the ways things can go wrong: the Security meta plugin and the Tunnel meta plugin. #### Security, mapped to your actual pipeline Rather than one generic "security" toggle, VibeControls' Security plugin dispatches to a different provider at each lifecycle stage — from a developer's local pre-commit hook through pull request, build, release, and runtime monitoring. Today that includes providers for local secrets scanning (`gitleaks protect`), fast and deep SAST/SCA on pull requests (Semgrep, osv-scanner), SBOM generation and vulnerability scanning at build time (Syft, Grype), dynamic scanning against preview environments using [OWASP ZAP](https://www.zaproxy.org/), OpenSSF Scorecard checks, signing and provenance at package publish (Cosign, SLSA), an OPA-backed release gate before production, and continuous runtime checks against live workloads. This stage-by-stage design isn't just a nice-to-have. Software bills of materials (SBOMs) in particular have gone from a niche practice to something procurement teams now expect: in Linux Foundation research on SBOM adoption, roughly nine in ten surveyed organizations said they had already started or were actively planning SBOM adoption ([Linux Foundation, "The State of Software Bill of Materials (SBOM) and Cybersecurity Readiness"](https://www.linuxfoundation.org/research/the-state-of-software-bill-of-materials-sbom-and-cybersecurity-readiness)). Treating security as one plugin per stage, rather than one blanket scan, is what makes it possible to meet that expectation without slowing every commit down with every check. #### Tunnels, without the manual `ssh -L` juggling The Tunnel meta plugin routes traffic across whichever provider you register: - **Cloudflare Tunnel** — exposes a localhost endpoint through `cloudflared` with named tunnels, useful when you need a public HTTPS URL for a demo or webhook receiver - **VibeTunnels** — an frp-based provider for self-hostable, air-gapped relays where you can't or won't route traffic through a third-party edge Both are governed the same way as everything else on the platform: RBAC-checked and audited, so a tunnel opened for a demo doesn't turn into a forgotten, unmonitored door left open on a server. ### How to Choose the Right Plugins for Your Team With more than 50 plugins across seven categories, the temptation is to install everything. Don't. A few questions narrow the list quickly: - **Which AI providers does your team actually pay for?** Install those AI provider plugins first — Claude, Codex, Gemini, Copilot, whatever's already in your budget — and add more only when a project genuinely needs a second option. - **Do you work solo or does state need to be shared?** One developer, one machine: Skalex. A team sharing session history or credentials across agents: Postgres. - **What does your security pipeline already require?** Match provider plugins to stages you already gate on — secrets scanning and PR-time SAST are usually the first two most teams want, with SBOM and release-gate providers added as compliance needs grow. - **Do you need to expose anything publicly?** Only install a tunnel provider if you actually need to share a local service; it's one less thing to audit if you don't. - **Which terminal do you already use?** Pick the session provider that matches your muscle memory (tmux is the most common starting point) rather than switching multiplexers just because a plugin exists. ### Building Your Own Plugin If none of the published plugins fit — a niche AI provider, an internal security scanner, a proprietary storage backend — the Plugin SDK is published and MIT licensed on GitHub, the same as the plugins it powers. Plugins declare a manifest, implement lifecycle hooks, and can expose their own REST routes and UI panels. You can keep a custom plugin private to your organization or publish it for others to install. ### Frequently Asked Questions **What is VibeControls?** VibeControls is an Agentic Engineering IDP (Internal Developer Platform) for AI-native teams. It runs a lightweight agent on every machine you work on — laptop, remote server, container, or cloud VM — and connects them into one control plane for sessions, tunnels, AI orchestration, and automation. **Are VibeControls plugins open source?** Yes. The plugin ecosystem — over 50 packages spanning AI, security, GitOps, session, storage, and tunnel providers — is MIT licensed and published on GitHub. The core agent, backend, CLI, and web app are closed source. **Do I need to install every plugin?** No. The agent ships small on purpose. Install the AI providers you actually use, the session multiplexer you already work in, and the security or tunnel providers your workflow requires — skip the rest. **Can I run more than one AI provider at once?** Yes. Install multiple `vibe-plugin-ai-*` provider plugins and choose which one drives a given session. This is useful for teams that want to compare models or that are contractually required to use a specific vendor on certain projects. **What's the difference between a meta plugin and a provider plugin?** A meta plugin (like AI or Security) defines a single interface. A provider plugin (like the Claude or GitHub provider) implements that interface for one specific backend. Installing a new provider never requires changing how you interact with the meta plugin. **Can I build a plugin that isn't in the catalogue?** Yes. The Plugin SDK is open source, and plugins declare a manifest, lifecycle hooks, and optional UI panels. Publish it for others to use or keep it private to your team. ### Next Steps The full, searchable catalogue — every meta, provider, and tool plugin, filterable by category — lives on the [VibeControls plugins page](https://vibecontrols.com/plugins). To see how plugins fit into the rest of the platform, from agents and tunnels to team RBAC, visit [VibeControls features](https://vibecontrols.com/features), or browse real [use cases](https://vibecontrols.com/use-cases) to see which plugin combinations fit workflows like yours. Have a question this guide didn't cover? Check the full [FAQ](https://vibecontrols.com/faq). VibeControls is in the run-up to launch. If you want early access when it opens, [join the waitlist](https://vibecontrols.com/waitlist) — you'll be first in line to install your first plugin. --- # VibeControls Cost Calculator > Compare VibeControls flat-rate pricing to credit-based and per-hour competitors. Transparent estimates for your team size and usage. Source: https://vibecontrols.com/calculator ## Flat-rate savings, in numbers Compare VibeControls against the per-hour, per-credit, and per-seat pricing models you are already evaluating. Adjust team size and usage to see the difference. VibeControls Team pricing starts at 5 seats (annual). ### Estimated monthly cost Competitor estimates are derived from the published /vs comparison data. | Platform | Model | Monthly estimate | vs VibeControls | | --- | --- | --- | --- | | Warp | Credit-based AI terminal. Build $20/mo, Max $20–$200/mo. | $200 | -$90 | | Termius | SSH vault. Team $20/user/mo. No runtime or AI included. | $200 | -$90 | | VibeControls Flat rate | Flat, regional per-seat pricing. Every feature on every plan. | $290 | — | | Port | SaaS IDP. Basic $30/seat/mo for 50-seat packages. | $300 | +$10 | | GitHub Codespaces | Per-hour compute + storage. ~$29–$115/mo full-time 4-core. | $720 | +$430 | | Backstage (Roadie managed) | Managed OSS IDP. Roadie ~$18K–$30K/yr + seats. | $1,750 | +$1,460 | #### Why flat rate wins at scale VibeControls bundles agents, sessions, tunnels, AI orchestration, and an initial IDP for one predictable per-seat price. Competitors charge per-hour compute (Codespaces), credits (Warp), or require a separate managed IDP contract (Port, Backstage) on top of the terminal tool. The calculator uses the midpoint of published ranges; your actual competitor spend may vary based on exact usage. ### Ready for predictable pricing? Start free, then scale through Hobby, Pro, and Team. Every feature on every plan. --- # Success Stories | VibeControls > Real-world stories and measurable outcomes from teams building with VibeControls. Source: https://vibecontrols.com/case-studies ## Case Studies Real-world stories of teams building with VibeControls. #### How a 200-engineer team scaled with VibeControls A fast-growing engineering org standardized on VibeControls and cut cycle time in half. #### From pilot to production in 30 days with VibeControls A regulated enterprise took VibeControls from a small pilot to company-wide rollout in a month. ### Want results like these? Tell us about your goals and we’ll show you exactly how VibeControls can help. --- # Changelog — VibeControls > Latest updates, improvements, and announcements for VibeControls. Source: https://vibecontrols.com/changelog ## Changelog Platform updates, improvements, and announcements — newest first. June 10, 2026 · The VibeControls Team ### VibeControls — June 2026 Release This release focuses on speed and polish across the platform. #### Added - New onboarding flow for first-time teams - Export to CSV/JSON from every list view #### Improved - Up to 40% faster agent orchestration - Refreshed navigation and dark-mode contrast #### Fixed - Edge cases in secure tunnels when working offline May 20, 2026 · The VibeControls Team ### VibeControls — May 2026 Release A feature-packed month for VibeControls. #### Added - Major upgrade to terminal sessions - Role-based access controls at every layer #### Improved - Clearer error messages across the product April 15, 2026 · The VibeControls Team ### VibeControls — Launch Foundations We laid the foundations for VibeControls — agentic engineering mission control. #### Added - Core agent orchestration - secure tunnels and terminal sessions - Multi-tenant isolation and audit logging --- # Contact Us - VibeControls > Get in touch with the VibeControls team for sales inquiries, technical support, or partnership opportunities. Source: https://vibecontrols.com/contact ## Get in Touch Have questions about VibeControls? We're here to help. Reach out to our team and we'll get back to you as soon as possible. ### Contact Information #### Email contact@vibecontrols.com #### Office Plot No.43 Veeramani Nagar 2nd Cross Street Nanmangalam, Chennai - 600117 #### Response Time We typically respond within 24 hours ### Send us a Message --- # Ecosystem — Apps, CLI, SDKs & MCP | VibeControls > Download VibeControls for desktop, install the CLI, Node/Python SDKs, and MCP server — plus the VS Code and Chrome extensions. Every surface in one place. Source: https://vibecontrols.com/ecosystem ## The VibeControls ecosystem Desktop apps, a CLI, SDKs for Node & Python, an MCP server, and editor & browser extensions — download what you need or copy a one-line install. Everywhere you and your agents work. Install the agent in seconds Then run `vibe start` — your machine shows up as a live, managed agent. ### Desktop App Native mission control for macOS, Linux & Windows. ### CLI · SDKs · MCP Install the agent, CLI, SDKs and MCP server from npm & PyPI. ### Editor & Browser VibeControls inside VS Code and your browser. ### Mobile VibeControls on the go — coming soon. ### Ready to put it all together? Spin up a workspace, connect an agent, and start orchestrating in minutes. --- # Desktop App — VibeControls Ecosystem > Access VibeControls in a native desktop app. Manage agents, vibes, and AI in one window. Choose your platform and download the latest version. Source: https://vibecontrols.com/ecosystem/desktop ## Desktop App Native mission control for macOS, Linux & Windows. Access VibeControls in a native desktop app. Manage agents, vibes, and AI in one window. Choose your platform and download the latest version. Builds are signed and auto-update. Prefer the terminal? [Install the CLI or agent instead](https://vibecontrols.com/ecosystem/packages). ### Explore the rest of the ecosystem --- # Editor & Browser — VibeControls Ecosystem > Bring agents, vibes, sessions and tunnels into the tools you already live in — your editor and your browser. Source: https://vibecontrols.com/ecosystem/extensions ## Editor & Browser VibeControls inside VS Code and your browser. Bring agents, vibes, sessions and tunnels into the tools you already live in — your editor and your browser. ### VS Code Extension Agents, vibes, sessions, tunnels & notes in your editor sidebar. The VibeControls — Agentic IDP extension adds a full sidebar for your agents, vibes, sessions, targets, tunnels and notes, right inside VS Code. Install from the Marketplace or via the command line. Or from the command line: ### Chrome Extension Open the VibeControls side panel from any tab. A browser side panel for quick access to your agents and sessions while you work. Launching on the Chrome Web Store soon. ### Explore the rest of the ecosystem --- # Mobile — VibeControls Ecosystem > Native mobile apps to check on agents, approve actions and keep an eye on your environments from anywhere. In active development. Source: https://vibecontrols.com/ecosystem/mobile ## Mobile VibeControls on the go — coming soon. Native mobile apps to check on agents, approve actions and keep an eye on your environments from anywhere. In active development. ### iOS App Native iPhone & iPad app. Monitor agents, sessions and tunnels and approve actions from your iPhone or iPad. ### Android App Native Android app. Monitor agents, sessions and tunnels and approve actions from your Android device. ### Explore the rest of the ecosystem --- # CLI · SDKs · MCP — VibeControls Ecosystem > Automate VibeControls with CLI, SDKs, and MCP server. Script and embed functionalities using TypeScript, Python, and control-plane tools. Source: https://vibecontrols.com/ecosystem/packages ## CLI · SDKs · MCP Install the agent, CLI, SDKs and MCP server from npm & PyPI. Automate VibeControls with CLI, SDKs, and MCP server. Script and embed functionalities using TypeScript, Python, and control-plane tools. ### Agent Local daemon + `vibe` CLI that turns any machine into a managed node. A lightweight local service for tmux sessions, SSH, cloudflared tunnels, port forwards, background tasks and a plugin system. Install it globally and run `vibe start` — your machine shows up as a live, managed agent in seconds. ### CLI Control-plane client `vibecontrols` — manage your whole team from the terminal. The platform CLI talks to the public GraphQL gateways with dual-token auth. Run `vibecontrols auth login`, then drive agents, vibes, sessions and tunnels across your team with a full subcommand suite. Separate from the agent: the agent runs locally on each machine, the CLI is your control-plane client. ### Node SDK Official TypeScript SDK for programmatic access to the full API. Typed, promise-based access to agents, vibes, sessions, tunnels and more. Drop it into any Node or Bun project to automate VibeControls from your own code. ### Python SDK Async Python SDK built on httpx — 14 domain modules. A fully async SDK with a dual-endpoint GraphQL client, multiple auth flows, automatic token refresh and retry-with-backoff. Ideal for data pipelines, automation and notebooks. ### MCP Server Wire VibeControls into Claude, Cursor & any Model Context Protocol client. A Model Context Protocol server exposing VibeControls tools, resources and prompts to AI assistants. Add the block below to your client config (e.g. `.mcp.json` or your editor’s MCP settings) and your agent can drive VibeControls directly. ``` { "mcpServers": { "vibecontrols": { "command": "npx", "args": ["-y", "@vibecontrols/mcp"] } } } ``` ### Explore the rest of the ecosystem --- # Explore Burdenoff products with AI — VibeControls > Ask anything about VibeControls and every Burdenoff product. Describe a problem and find the use case that solves it, with links to the pages that explain it. Source: https://vibecontrols.com/explore ## Ask anything about VibeControls Ask about features, use cases or pricing for VibeControls — or compare it with anything else we build. Every answer comes straight from our documentation. - Grounded in our own docs Every answer comes from published product pages and documentation — never invented. - Compare across the ecosystem Ask which product fits your team, or how two of them work together. - Straight to the right page Get the sources and next steps — pricing, docs, a demo — without hunting. Try asking Focus on a product --- # FAQ — VibeControls > Find answers to common questions about VibeControls, including agents, sessions, AI, security, pricing, and extensibility. Source: https://vibecontrols.com/faq ## Frequently asked questions Everything you need to know about VibeControls — agents, sessions, vibes, tunnels, AI, security, and more. Can't find your answer? [Contact us](https://vibecontrols.com/contact). ### Frequently asked questions #### What is VibeControls? #### What is the official tagline? #### Who is VibeControls for? #### Is VibeControls open source? #### How do I install the agent on my machine? #### What is the difference between the agent and the CLI? #### Can I run multiple agents on one machine? #### How does the agent stay alive if I close my terminal? #### Can I install an agent remotely on an SSH server? #### What session types are supported? ### Still have questions? Our team is happy to help — reach out via contact form or explore the docs. --- # Modern Development Features | VibeControls > Features grouped into Core Infrastructure, Interface & Tooling, and Team & Ops — agents, tunnels, sandboxes, terminals, code-server, plugins, AI, RBAC, audit. Source: https://vibecontrols.com/features ## Powerful Features for Modern Development Discover everything that makes VibeControls the ultimate mission control for developers and teams. ### Core Infrastructure Agents on every machine, secure tunnels between them, sandboxes for safe execution. #### Vibe Agents Every machine, one console Lightweight agents on local, SSH, container, or cloud machines. Secure real-time control plane. SQLite local state survives offline. Register a target in seconds, never lose context. - macOS, Linux, Windows, containers, cloud VMs - WebSocket real-time + REST control plane - SQLite local state for offline resilience - Auto-reconnect with health continuously reported - Plugins for tmux, SSH, Git, containers, AI providers #### Secure Tunnels Port forwarding, governed Local-to-remote, remote-to-local, dynamic SOCKS, or named Cloudflare tunnels — all from a visual interface, all RBAC-checked, all audited. - Local-to-remote port forwarding - Remote-to-local service exposure - Dynamic SOCKS proxy + Cloudflare tunnels - Connection health + auto-reconnect - Encrypted credential storage (AES-256) #### Sandboxes Disposable, reproducible envs Spin up isolated dev environments backed by Cloudflare Sandboxes — perfect for AI agent execution, untrusted scripts, or "try-before-merge" PR previews. Clean teardown when you're done. - Cloudflare-Sandbox backed isolation - Resource-limited containers - Snapshot, fork, and restore - Template-based provisioning - Per-vibe filesystem export to files-svc ### Interface & Tooling How developers actually drive the system — terminals in the browser, code-server, plugin ecosystem, AI workflows. #### Terminal Sessions Persistent, streamable, shareable tmux, SSH, and custom terminal sessions with real-time output streaming to the browser. Attach, detach, share live with a teammate, and search through full history. - tmux persistent sessions - SSH multiplexing + jump-host support - Real-time output streaming over WS - Searchable session history + replay - Live co-watch with view/edit RBAC #### Code Server Edit code from anywhere Monaco-based code editor in your browser with syntax highlighting and IntelliSense. Edit files on any connected agent without leaving VibeControls. - Monaco editor (VS Code engine) - Syntax highlighting for 100+ languages - IntelliSense and autocomplete - File-system navigation - Multi-tab editing #### Plugin Ecosystem 50+ plugins, three-tier model Meta plugins own the interface, provider plugins implement it for specific backends (tmux, wezterm, Cloudflare, Claude…). Standalone tool plugins fill the rest. Build your own with the SDK. - Database GUI, cache GUI, container manager - Monaco code editor, Git, API client - Meta / provider / tool plugin tiers - Sign-and-pin install via sha256 - Build & publish your own with the Plugin SDK #### AI Assistant & AI Builder BYOK, governed, cited One interface, many providers — Anthropic Claude, OpenAI Codex, Gemini, Copilot, Aider, OpenCode, Plandex, any OpenAI-compatible endpoint. You bring the key. Prompts are governed, logged, and cited. - 12+ providers behind one interface - BYOK — your keys, your inference bill - Plan mode, code mode, and ask mode - Cited responses with file/line provenance - MCP server for Claude Desktop & Claude Code ### Team & Ops Make the platform safe to share — automation, RBAC, audit, scheduling, collaboration. #### Workflow Automation & Scheduling Cron, webhooks, AI triggers Cron-style scheduled tasks running on the agent. Vibe Deck — one-click button grids that fire scripts, webhooks, or AI prompts. Replace half your shell aliases. - Cron-style scheduled tasks per vibe - Vibe Deck — button grid for scripts, commands, webhooks - AI-prompt + agent-action triggers - Trigger workflows from calendar events - Per-vibe runbooks and notes ### More Capabilities Additional features that make VibeControls a complete development platform. #### SSH Management Visual SSH connection manager with encrypted credential storage, jump host support, and auto-reconnection. #### AI Integration Built-in Claude Code integration and AI-powered coding assistant via the plugin system. #### Enterprise Security AES-256-GCM at rest, TLS in transit, gateway-enforced RBAC, SSO + SCIM (Enterprise), immutable 7-year audit logging. #### Project Notes Markdown-formatted project notes with tagging, full-text search, and team sharing. #### Prompt Templates Reusable AI prompts with variables, categories, team sharing, and usage analytics. #### Multi-Platform Works on macOS, Linux, Windows, containers, and cloud instances. One platform, every environment. ### Team Collaboration VibeControls is built for teams. Share vibes with granular permissions, collaborate on terminal sessions, and manage team workspaces with centralized billing and quotas. #### Sharing - View / Edit / Admin permissions - Shared SSH connections - Team-wide plugin installs - Expiration controls #### Management - Role-based access control - Centralized billing - Quota management - Team workspaces ### Ready to experience these features? Start using VibeControls today and transform how you manage your development environment. --- # Cookie Policy | VibeControls > Learn how VibeControls uses cookies. Source: https://vibecontrols.com/legal/cookies ## Cookie Policy Last updated: May 16, 2026 ### 1. What Are Cookies Cookies are small text files stored on your device to help us provide a better experience. ### 2. Types of Cookies #### Essential Cookies Required for authentication, session management, and security. Cannot be disabled. #### Analytics Cookies Help us understand usage patterns. We use privacy-focused analytics (Rybbit). #### Preference Cookies Remember your settings like theme preference and language. ### 3. How to Control Cookies You can control cookies through your browser settings. ### 4. Contact Email: contact@vibecontrols.com --- # Cookie Preferences — VibeControls > Choose which cookies VibeControls may use. Change your analytics, functional, and marketing choices at any time. Source: https://vibecontrols.com/legal/cookies/preferences Legal ## Cookie preferences Control how VibeControls uses cookies and similar technologies. Strictly necessary cookies keep the site working and can't be turned off. For the full detail, read our [Cookie Policy](https://vibecontrols.com/legal/cookies). ### Strictly necessary Required to operate the site — security, load balancing, and remembering your cookie choices. Always on. ### Functional Remembers preferences such as theme and language so the site behaves the way you set it up. ### Analytics & performance Lets us measure how the site is used (via Rybbit) so we can fix bugs and prioritise improvements. We do not sell this data. ### Marketing Measures the effectiveness of our own marketing. We never share data with third-party advertising networks for cross-context behavioural advertising. --- # Privacy Policy | VibeControls > Learn how VibeControls collects, uses, and protects your personal data. Source: https://vibecontrols.com/legal/privacy ## Privacy Policy Last updated: May 16, 2026 ### 1. Introduction Welcome to VibeControls ("we," "our," or "us"), developed by Algoshred Technologies Pvt. Ltd. We respect your privacy and are committed to protecting your personal data. ### 2. Information We Collect #### 2.1 Information You Provide - Account information (name, email, username) - Profile information and preferences - Project data and configurations - Communication with our support team #### 2.2 Information We Collect Automatically - Usage data and analytics - Device information (IP address, browser type) - Session information and access logs ### 3. How We Use Your Information - Provide and maintain our services - Improve and optimize our platform - Communicate about updates and support - Ensure security and prevent fraud - Comply with legal obligations ### 4. Data Storage and Security We implement appropriate security measures to protect your personal information. Your data is encrypted both in transit and at rest. ### 5. Your Rights - Access your personal information - Correct inaccurate data - Request deletion of your data - Export your data - Opt out of marketing communications ### 6. Contact Email: contact@vibecontrols.com | Address: Plot No.43, Veeramani Nagar, Chennai - 600117, India *© 2026 Algoshred Technologies Pvt. Ltd. All rights reserved.* --- # Refund & Cancellation Policy | VibeControls Source: https://vibecontrols.com/legal/refunds ## Refund & Cancellation Policy ```

Content is being updated. Please check back later.

``` --- # Terms of Service - VibeControls > Terms and conditions for using VibeControls platform. Source: https://vibecontrols.com/legal/terms ## Terms of Service Last updated: May 16, 2026 ### 1. Acceptance of Terms By accessing VibeControls, developed by Algoshred Technologies Pvt. Ltd., you agree to these terms. ### 2. Description of Service VibeControls provides centralized control over development projects, terminals, and remote environments. ### 3. Acceptable Use - Do not use for unlawful purposes - Do not attempt unauthorized access - Do not interfere with the Service - Do not reverse engineer the Service ### 4. Intellectual Property VibeControls and all related content are owned by Algoshred Technologies Pvt. Ltd. ### 5. Limitation of Liability To the maximum extent permitted by law, we shall not be liable for indirect, incidental, or consequential damages. ### 6. Termination We may terminate or suspend access for breach of these Terms. ### 7. Contact Email: contact@vibecontrols.com *© 2026 Algoshred Technologies Pvt. Ltd. All rights reserved.* --- # Invalid Unsubscribe Link | VibeControls plugin > Manage your newsletter subscription preferences. Source: https://vibecontrols.com/newsletter/unsubscribe ## Invalid Unsubscribe Link This unsubscribe link is invalid or missing. If you want to unsubscribe, please use the link from your newsletter email. --- # Partner Programs | VibeControls > Join the VibeControls Partner Program — six tracks with commission protection, training, and co-marketing. Download the Partner / Reseller Agreement and apply. Source: https://vibecontrols.com/partners ## Partner with VibeControls Refer, resell, consult, or build alongside us — six tracks spanning Referral, Consulting, Implementation, Reseller, MSP, and Strategic Alliance. Join a growing ecosystem of consultants, resellers, and technology partners powering developer productivity at scale. ### Partner tracks We offer **six tracks** for partners who want to grow with VibeControls: - **Referral** — Earn commissions for sending qualified opportunities our way. Light-touch, no certification required. - **Consulting** — Implement, train, and customise VibeControls for end customers. Certified consultants only. - **Implementation** — Deep technical engagement on enterprise deployments. Trained on workspace architecture, RBAC, plugin development. - **Authorised Reseller** — Resell VibeControls licences in your territory. Requires a signed Partner Schedule. - **Managed Service Provider (MSP)** — Operate VibeControls deployments on behalf of end customers. - **Strategic Alliance** — Co-sell, co-market, and integrate with our platform. ### Why partner? - **Registered opportunity protection** — Default 120-day exclusivity window on registered deals - **Commission / margin** — Set by Partner Schedule and tier (see the Commission Guide) - **Co-branded marketing** — Use approved VibeControls assets, joint case studies, and event support - **Training and certification** — Free for active partners, certified specialisations available - **Dedicated partner manager** — From qualifying tier upward ### How to join - **Read the Partner / Reseller Agreement** — download the full agreement below. - **Submit your interest** — tell us about your organisation using the form at the bottom of this page. - **Sign a Partner Schedule** — we'll send you a tier-specific schedule based on your model. - **Get onboarded** — certifications, portal access, joint go-to-market planning. We respond to qualified partner inquiries within **2 business days**. ### Already a partner? Sign in to the [Partner Portal](https://app.vibecontrols.com) to register deals, access training, and view commission statements. ### Partner Documents Download the one-page cheat sheet, the full partner playbook, and the Burdenoff product portfolio — plus the programme terms — before applying. #### Partner / Reseller Agreement The full VibeControls Partner / Reseller Agreement — review the program terms before applying. #### VibeControls Partner Cheat Sheet Version 1.0 One page: what VibeControls is, who buys it, what it costs, and how to introduce us. #### VibeControls Partner Playbook Version 1.0 The full brief: positioning, buyer personas, use cases, pricing, competitive landscape, objection handling, lead handoff and partner commercials — plus every product screen. #### Burdenoff Product Portfolio Version 1.0 Every Burdenoff product on one sheet, with the signals that tell you which one a conversation is pointing at. ### Contact Information #### Email contact@vibecontrols.com #### Office Plot No.43 Veeramani Nagar 2nd Cross Street Nanmangalam, Chennai - 600117 #### Response Time We typically respond within 24 hours ### Send us a Message --- # AI Plugins Catalogue | VibeControls > Explore VibeControls plugins: AI, security, tunneling, sessions, storage, GitOps, and developer workflows for enhanced productivity. Source: https://vibecontrols.com/plugins ## VibeControls plugin catalogue Meta interfaces, concrete providers, standalone tools — the agent ships small. Install what you need. Build your own with the Plugin SDK. Every plugin and the Plugin SDK are open source under the MIT License. The`@vibecontrols/agent`runtime that loads them is closed-source. ### Meta plugins Each meta plugin owns one interface contract. Providers below implement that contract for a specific backend. Standalone tools sit alongside the hierarchy. #### AI AI tool orchestrator — one interface, many providers (Claude, Codex, Gemini, Copilot, Aider, OpenCode, Plandex, OpenAI-compatible). #### Security Security lifecycle orchestrator — dispatches every dev / pr / merge / build / release stage to the right per-stage security provider. #### GitOps GitOps orchestrator — one interface for GitHub, GitLab, Bitbucket Cloud, and Azure DevOps repo / PR / CI / security queries. #### Plan Plan orchestration meta-plugin — owns the PlanProvider contract and dispatches to plan-mode providers like Plannotator. #### Storage Storage meta plugin — owns the AgentDatabase contract so embedded and shared backends are interchangeable behind one API. #### Session Manager Unified session manager — capability discovery, feature negotiation, and provider routing across tmux, WezTerm, and Zellij. #### Tunnel Unified tunnel manager — routes requests across registered tunnel providers (Cloudflare, VibeTunnels). ### All plugins (56) #### AI AI tool orchestrator — one interface, many providers (Claude, Codex, Gemini, Copilot, Aider, OpenCode, Plandex, OpenAI-compatible). #### Security Security lifecycle orchestrator — dispatches every dev / pr / merge / build / release stage to the right per-stage security provider. #### GitOps GitOps orchestrator — one interface for GitHub, GitLab, Bitbucket Cloud, and Azure DevOps repo / PR / CI / security queries. #### Plan Plan orchestration meta-plugin — owns the PlanProvider contract and dispatches to plan-mode providers like Plannotator. #### Storage Storage meta plugin — owns the AgentDatabase contract so embedded and shared backends are interchangeable behind one API. #### Session Manager Unified session manager — capability discovery, feature negotiation, and provider routing across tmux, WezTerm, and Zellij. #### Tunnel Unified tunnel manager — routes requests across registered tunnel providers (Cloudflare, VibeTunnels). #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. #### Gemini Google Gemini CLI provider — Gemini Pro 1.5 / 2.0 ready. #### Goose Block Goose agent provider for VibeControls. #### Minimax Minimax CLI + Anthropic-compatible SDK provider. #### Ollama Ollama (local + cloud) — self-hosted models, BYOK API key. #### OpenAI-compatible Any OpenAI-compatible endpoint — point it at vLLM, LM Studio, Bedrock proxies, etc. #### OpenCode OpenCode CLI agent provider. #### OpenRouter OpenRouter SDK provider — pay-as-you-go to 100+ models. #### Pi Pi.ai conversational agent provider. #### Plandex Plandex multi-step AI engineering agent. #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. #### Release gate OPA-backed release-gate at promote.prod — block on policy, evidence-required deploys. #### Scheduled rescan Nightly Grype offline rescan against last-built SBOMs. #### Runtime continuous Trivy + kube-bench continuous runtime checks against live workloads. #### SAST deep Full SAST + SCA sweep (Semgrep + osv-scanner) on pull_request.deep. #### SBOM build Syft CycloneDX SBOM + Grype scan at build lifecycle stage. #### OpenSSF Scorecard OpenSSF Scorecard checks at main.merge — track posture over time. #### Secrets (PR) Gitleaks-backed secret scanner for pull_request.fast — fail fast on PR. #### GitHub GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. #### GitLab GitLab provider for the GitOps meta plugin — REST v4. #### Bitbucket Cloud Bitbucket Cloud provider for the GitOps meta plugin — REST v2. #### Azure DevOps Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. #### tmux Tmux + ttyd session provider — the default for persistent terminal sessions. #### WezTerm WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. #### Zellij Zellij + ttyd session provider — terminal workspace with floating panes. #### Skalex Document-oriented file-backed storage provider with AES-256-GCM at rest. #### Postgres Full-parity shared database storage provider with AES-256-GCM encryption — share state across multiple agents. #### Cloudflare Tunnel Cloudflare Tunnel provider — expose localhost endpoints via cloudflared with named tunnels. #### VibeTunnels frp-based tunnel provider — self-hostable tunnel relay for air-gapped networks. #### Plannotator Plannotator plan provider — wraps the upstream CLI and configures per-AI-agent hooks. #### AI config sync Cross-harness AI configuration management via Bridle — sync prompts + settings. #### Backup Solution | VibeControls plugin Back up Skalex-backed agent state to S3 or any custom storage — restore on demand. #### code-server Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. #### Git client Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. #### GraphiQL API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. #### SSH SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. #### UI Hello Minimal example UI plugin — copy as a starting point for new iframe-mounted plugin UIs. --- # AI config sync — VibeControls plugin > Cross-harness AI configuration management via Bridle — sync prompts + settings. Source: https://vibecontrols.com/plugins/agent-aiconfig ## AI config sync for VibeControls Cross-harness AI configuration management via Bridle — sync prompts + settings. ### About this plugin > Cross-harness AI configuration management via Bridle — sync prompts + settings. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-agent-aiconfig ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-agent-aiconfig # or npm install @vibecontrols/vibe-plugin-agent-aiconfig ``` ### How it works **Agent** plugins extend the agent itself with cross-cutting capabilities (state backup / restore, AI-provider configuration, etc.). They are not tied to a single meta plugin contract. ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-agent-aiconfig](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-agent-aiconfig) - Source: [https://github.com/algoshred/vibe-plugin-agent-aiconfig](https://github.com/algoshred/vibe-plugin-agent-aiconfig) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/agent-aiconfig](https://vibecontrols.com/plugins/agent-aiconfig) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-agent-aiconfig` - List installed plugins: `vibe plugin list` ### Related Agent plugins #### Backup Solution | VibeControls plugin Back up Skalex-backed agent state to S3 or any custom storage — restore on demand. --- # Backup Solution | VibeControls plugin — VibeControls plugin > Back up Skalex-backed agent state to S3 or any custom storage — restore on demand. Source: https://vibecontrols.com/plugins/agent-backup ## Backup Solution | VibeControls plugin for VibeControls Back up Skalex-backed agent state to S3 or any custom storage — restore on demand. ### About this plugin > Back up Skalex-backed agent state to S3 or any custom storage — restore on demand. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-agent-backup ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-agent-backup # or npm install @vibecontrols/vibe-plugin-agent-backup ``` ### How it works **Agent** plugins extend the agent itself with cross-cutting capabilities (state backup / restore, AI-provider configuration, etc.). They are not tied to a single meta plugin contract. ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-agent-backup](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-agent-backup) - Source: [https://github.com/algoshred/vibe-plugin-agent-backup](https://github.com/algoshred/vibe-plugin-agent-backup) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/agent-backup](https://vibecontrols.com/plugins/agent-backup) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-agent-backup` - List installed plugins: `vibe plugin list` ### Related Agent plugins #### AI config sync Cross-harness AI configuration management via Bridle — sync prompts + settings. --- # AI plugins — VibeControls > AI tool orchestrator — one interface, many providers (Claude, Codex, Gemini, Copilot, Aider, OpenCode, Plandex, OpenAI-compatible). 16 providers available. Source: https://vibecontrols.com/plugins/ai ## AI plugins for VibeControls AI tool orchestrator — one interface, many providers (Claude, Codex, Gemini, Copilot, Aider, OpenCode, Plandex, OpenAI-compatible). AI providers behind one orchestration interface. Claude, Codex, Gemini, Copilot, Aider — bring your key, keep your prompts. ### 16 providers Each provider implements the AI contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. #### Gemini Google Gemini CLI provider — Gemini Pro 1.5 / 2.0 ready. #### Goose Block Goose agent provider for VibeControls. #### Minimax Minimax CLI + Anthropic-compatible SDK provider. #### Ollama Ollama (local + cloud) — self-hosted models, BYOK API key. #### OpenAI-compatible Any OpenAI-compatible endpoint — point it at vLLM, LM Studio, Bedrock proxies, etc. #### OpenCode OpenCode CLI agent provider. #### OpenRouter OpenRouter SDK provider — pay-as-you-go to 100+ models. #### Pi Pi.ai conversational agent provider. #### Plandex Plandex multi-step AI engineering agent. --- # Aider — VibeControls plugin > Aider CLI agent — pair programmer that edits your local git tree. Source: https://vibecontrols.com/plugins/ai-aider ## Aider for VibeControls Aider CLI agent — pair programmer that edits your local git tree. ### About this plugin > Aider CLI agent — pair programmer that edits your local git tree. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-aider ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-aider # or npm install @vibecontrols/vibe-plugin-ai-aider ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Aider** — [https://github.com/Aider-AI/aider](https://github.com/Aider-AI/aider) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-aider](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-aider) - Source: [https://github.com/algoshred/vibe-plugin-ai-aider](https://github.com/algoshred/vibe-plugin-ai-aider) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-aider](https://vibecontrols.com/plugins/ai-aider) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-aider` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # Amp — VibeControls plugin > Sourcegraph Amp — agentic IDE actions backed by your codebase index. Source: https://vibecontrols.com/plugins/ai-amp ## Amp for VibeControls Sourcegraph Amp — agentic IDE actions backed by your codebase index. ### About this plugin > Sourcegraph Amp — agentic IDE actions backed by your codebase index. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-amp ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-amp # or npm install @vibecontrols/vibe-plugin-ai-amp ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Sourcegraph Amp** — [https://ampcode.com/](https://ampcode.com/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-amp](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-amp) - Source: [https://github.com/algoshred/vibe-plugin-ai-amp](https://github.com/algoshred/vibe-plugin-ai-amp) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-amp](https://vibecontrols.com/plugins/ai-amp) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-amp` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # Claude — VibeControls plugin > Anthropic Claude Code agent — the default reference provider. Source: https://vibecontrols.com/plugins/ai-claude ## Claude for VibeControls Anthropic Claude Code agent — the default reference provider. ### About this plugin > Anthropic Claude Code agent — the default reference provider. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-claude ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-claude # or npm install @vibecontrols/vibe-plugin-ai-claude ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Claude Code** — [https://github.com/anthropics/claude-code](https://github.com/anthropics/claude-code) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-claude](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-claude) - Source: [https://github.com/algoshred/vibe-plugin-ai-claude](https://github.com/algoshred/vibe-plugin-ai-claude) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-claude](https://vibecontrols.com/plugins/ai-claude) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-claude` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # Codex — VibeControls plugin > OpenAI Codex CLI — agentic coding from OpenAI. Source: https://vibecontrols.com/plugins/ai-codex ## Codex for VibeControls OpenAI Codex CLI — agentic coding from OpenAI. ### About this plugin > OpenAI Codex CLI — agentic coding from OpenAI. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-codex ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-codex # or npm install @vibecontrols/vibe-plugin-ai-codex ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **OpenAI Codex CLI** — [https://github.com/openai/codex](https://github.com/openai/codex) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-codex](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-codex) - Source: [https://github.com/algoshred/vibe-plugin-ai-codex](https://github.com/algoshred/vibe-plugin-ai-codex) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-codex](https://vibecontrols.com/plugins/ai-codex) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-codex` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # GitHub Copilot — VibeControls plugin > GitHub Copilot CLI for the agent runtime. Source: https://vibecontrols.com/plugins/ai-copilot ## GitHub Copilot for VibeControls GitHub Copilot CLI for the agent runtime. ### About this plugin > GitHub Copilot CLI for the agent runtime. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-copilot ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-copilot # or npm install @vibecontrols/vibe-plugin-ai-copilot ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **GitHub Copilot CLI** — [https://github.com/github/copilot-cli](https://github.com/github/copilot-cli) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-copilot](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-copilot) - Source: [https://github.com/algoshred/vibe-plugin-ai-copilot](https://github.com/algoshred/vibe-plugin-ai-copilot) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-copilot](https://vibecontrols.com/plugins/ai-copilot) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-copilot` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # Charm Crush — VibeControls plugin > Charm Crush AI agent provider — TUI-native chat. Source: https://vibecontrols.com/plugins/ai-crush ## Charm Crush for VibeControls Charm Crush AI agent provider — TUI-native chat. ### About this plugin > Charm Crush AI agent provider — TUI-native chat. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-crush ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-crush # or npm install @vibecontrols/vibe-plugin-ai-crush ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Charm Crush** — [https://github.com/charmbracelet/crush](https://github.com/charmbracelet/crush) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-crush](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-crush) - Source: [https://github.com/algoshred/vibe-plugin-ai-crush](https://github.com/algoshred/vibe-plugin-ai-crush) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-crush](https://vibecontrols.com/plugins/ai-crush) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-crush` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Cursor Cursor CLI + SDK — bring your Cursor experience into the agent. --- # Cursor — VibeControls plugin > Cursor CLI + SDK — bring your Cursor experience into the agent. Source: https://vibecontrols.com/plugins/ai-cursor ## Cursor for VibeControls Cursor CLI + SDK — bring your Cursor experience into the agent. ### About this plugin > Cursor CLI + SDK — bring your Cursor experience into the agent. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-cursor ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-cursor # or npm install @vibecontrols/vibe-plugin-ai-cursor ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Cursor Agent** — [https://cursor.com/](https://cursor.com/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-cursor](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-cursor) - Source: [https://github.com/algoshred/vibe-plugin-ai-cursor](https://github.com/algoshred/vibe-plugin-ai-cursor) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-cursor](https://vibecontrols.com/plugins/ai-cursor) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-cursor` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Gemini — VibeControls plugin > Google Gemini CLI provider — Gemini Pro 1.5 / 2.0 ready. Source: https://vibecontrols.com/plugins/ai-gemini ## Gemini for VibeControls Google Gemini CLI provider — Gemini Pro 1.5 / 2.0 ready. ### About this plugin > Google Gemini CLI provider — Gemini Pro 1.5 / 2.0 ready. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-gemini ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-gemini # or npm install @vibecontrols/vibe-plugin-ai-gemini ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Gemini CLI** — [https://github.com/google-gemini/gemini-cli](https://github.com/google-gemini/gemini-cli) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-gemini](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-gemini) - Source: [https://github.com/algoshred/vibe-plugin-ai-gemini](https://github.com/algoshred/vibe-plugin-ai-gemini) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-gemini](https://vibecontrols.com/plugins/ai-gemini) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-gemini` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Goose — VibeControls plugin > Block Goose agent provider for VibeControls. Source: https://vibecontrols.com/plugins/ai-goose ## Goose for VibeControls Block Goose agent provider for VibeControls. ### About this plugin > Block Goose agent provider for VibeControls. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-goose ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-goose # or npm install @vibecontrols/vibe-plugin-ai-goose ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Block Goose** — [https://github.com/block/goose](https://github.com/block/goose) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-goose](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-goose) - Source: [https://github.com/algoshred/vibe-plugin-ai-goose](https://github.com/algoshred/vibe-plugin-ai-goose) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-goose](https://vibecontrols.com/plugins/ai-goose) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-goose` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Minimax — VibeControls plugin > Minimax CLI + Anthropic-compatible SDK provider. Source: https://vibecontrols.com/plugins/ai-minimax ## Minimax for VibeControls Minimax CLI + Anthropic-compatible SDK provider. ### About this plugin > Minimax CLI + Anthropic-compatible SDK provider. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-minimax ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-minimax # or npm install @vibecontrols/vibe-plugin-ai-minimax ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **MiniMax** — [https://www.minimax.io/](https://www.minimax.io/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-minimax](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-minimax) - Source: [https://github.com/algoshred/vibe-plugin-ai-minimax](https://github.com/algoshred/vibe-plugin-ai-minimax) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-minimax](https://vibecontrols.com/plugins/ai-minimax) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-minimax` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Ollama — VibeControls plugin > Ollama (local + cloud) — self-hosted models, BYOK API key. Source: https://vibecontrols.com/plugins/ai-ollama ## Ollama for VibeControls Ollama (local + cloud) — self-hosted models, BYOK API key. ### About this plugin > Ollama (local + cloud) — self-hosted models, BYOK API key. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-ollama ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-ollama # or npm install @vibecontrols/vibe-plugin-ai-ollama ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Ollama** — [https://github.com/ollama/ollama](https://github.com/ollama/ollama) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-ollama](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-ollama) - Source: [https://github.com/algoshred/vibe-plugin-ai-ollama](https://github.com/algoshred/vibe-plugin-ai-ollama) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-ollama](https://vibecontrols.com/plugins/ai-ollama) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-ollama` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # OpenAI-compatible — VibeControls plugin > Any OpenAI-compatible endpoint — point it at vLLM, LM Studio, Bedrock proxies, etc. Source: https://vibecontrols.com/plugins/ai-openai-compat ## OpenAI-compatible for VibeControls Any OpenAI-compatible endpoint — point it at vLLM, LM Studio, Bedrock proxies, etc. ### About this plugin > Any OpenAI-compatible endpoint — point it at vLLM, LM Studio, Bedrock proxies, etc. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-openai-compat ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-openai-compat # or npm install @vibecontrols/vibe-plugin-ai-openai-compat ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **OpenAI API Specification** — [https://platform.openai.com/docs/api-reference](https://platform.openai.com/docs/api-reference) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-openai-compat](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-openai-compat) - Source: [https://github.com/algoshred/vibe-plugin-ai-openai-compat](https://github.com/algoshred/vibe-plugin-ai-openai-compat) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-openai-compat](https://vibecontrols.com/plugins/ai-openai-compat) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-openai-compat` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # OpenCode — VibeControls plugin > OpenCode CLI agent provider. Source: https://vibecontrols.com/plugins/ai-opencode ## OpenCode for VibeControls OpenCode CLI agent provider. ### About this plugin > OpenCode CLI agent provider. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-opencode ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-opencode # or npm install @vibecontrols/vibe-plugin-ai-opencode ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **OpenCode** — [https://github.com/sst/opencode](https://github.com/sst/opencode) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-opencode](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-opencode) - Source: [https://github.com/algoshred/vibe-plugin-ai-opencode](https://github.com/algoshred/vibe-plugin-ai-opencode) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-opencode](https://vibecontrols.com/plugins/ai-opencode) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-opencode` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # OpenRouter — VibeControls plugin > OpenRouter SDK provider — pay-as-you-go to 100+ models. Source: https://vibecontrols.com/plugins/ai-openrouter ## OpenRouter for VibeControls OpenRouter SDK provider — pay-as-you-go to 100+ models. ### About this plugin > OpenRouter SDK provider — pay-as-you-go to 100+ models. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-openrouter ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-openrouter # or npm install @vibecontrols/vibe-plugin-ai-openrouter ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **OpenRouter** — [https://openrouter.ai/](https://openrouter.ai/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-openrouter](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-openrouter) - Source: [https://github.com/algoshred/vibe-plugin-ai-openrouter](https://github.com/algoshred/vibe-plugin-ai-openrouter) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-openrouter](https://vibecontrols.com/plugins/ai-openrouter) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-openrouter` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Pi — VibeControls plugin > Pi.ai conversational agent provider. Source: https://vibecontrols.com/plugins/ai-pi ## Pi for VibeControls Pi.ai conversational agent provider. ### About this plugin > Pi.ai conversational agent provider. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-pi ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-pi # or npm install @vibecontrols/vibe-plugin-ai-pi ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Pi (Inflection AI)** — [https://pi.ai/](https://pi.ai/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-pi](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-pi) - Source: [https://github.com/algoshred/vibe-plugin-ai-pi](https://github.com/algoshred/vibe-plugin-ai-pi) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-pi](https://vibecontrols.com/plugins/ai-pi) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-pi` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # Plandex — VibeControls plugin > Plandex multi-step AI engineering agent. Source: https://vibecontrols.com/plugins/ai-plandex ## Plandex for VibeControls Plandex multi-step AI engineering agent. ### About this plugin > Plandex multi-step AI engineering agent. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-ai-plandex ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-ai-plandex # or npm install @vibecontrols/vibe-plugin-ai-plandex ``` ### How it works AI **providers** implement the `AIProvider` contract from `@vibecontrols/vibe-plugin-ai` (meta). Once installed and registered, the agent can route prompts, hooks and MCP requests through this provider alongside any others you have configured. The meta plugin handles fan-out, provider selection and capability negotiation. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-ai` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Plandex** — [https://github.com/plandex-ai/plandex](https://github.com/plandex-ai/plandex) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-plandex](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-ai-plandex) - Source: [https://github.com/algoshred/vibe-plugin-ai-plandex](https://github.com/algoshred/vibe-plugin-ai-plandex) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/ai-plandex](https://vibecontrols.com/plugins/ai-plandex) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-ai` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ai-plandex` - List installed plugins: `vibe plugin list` ### Other AI plugins #### Aider Aider CLI agent — pair programmer that edits your local git tree. #### Amp Sourcegraph Amp — agentic IDE actions backed by your codebase index. #### Claude Anthropic Claude Code agent — the default reference provider. #### Codex OpenAI Codex CLI — agentic coding from OpenAI. #### GitHub Copilot GitHub Copilot CLI for the agent runtime. #### Charm Crush Charm Crush AI agent provider — TUI-native chat. --- # GitOps plugins — VibeControls > GitOps orchestrator — one interface for GitHub, GitLab, Bitbucket Cloud, and Azure DevOps repo / PR / CI / security queries. 4 providers available. Source: https://vibecontrols.com/plugins/gitops ## GitOps plugins for VibeControls GitOps orchestrator — one interface for GitHub, GitLab, Bitbucket Cloud, and Azure DevOps repo / PR / CI / security queries. One repo / PR / CI interface across GitHub, GitLab, Bitbucket, and Azure DevOps. ### 4 providers Each provider implements the GitOps contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### GitHub GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. #### GitLab GitLab provider for the GitOps meta plugin — REST v4. #### Bitbucket Cloud Bitbucket Cloud provider for the GitOps meta plugin — REST v2. #### Azure DevOps Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. --- # Azure DevOps — VibeControls plugin > Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. Source: https://vibecontrols.com/plugins/gitops-azdevops ## Azure DevOps for VibeControls Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. ### About this plugin Azure DevOps provider for the [`@vibecontrols/vibe-plugin-gitops`](https://npmjs.com/package/@vibecontrols/vibe-plugin-gitops) meta plugin. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-gitops # meta (once) vibe plugin install @vibecontrols/vibe-plugin-gitops-azdevops ``` ### Auth Personal Access Token (PAT) with these scopes ([https://dev.azure.com/{org}/_usersSettings/tokens](https://dev.azure.com/%7Borg%7D/_usersSettings/tokens)): - Code: read - Pull Request: read - Build: read - Release: read (optional) - Security: read (optional) ``` curl -X POST "${AGENT_URL}/api/profiles/default/gitops/azdevops/auth" \ -H "x-agent-api-key: $KEY" \ -d '{ "kind": "pat", "token": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx", "meta": { "organization": "my-org" } }' ``` The `meta.organization` field is required — Azure DevOps APIs are scoped per organisation. ### Repo FQN Azure DevOps FQN format: `{organization}/{project}/{repo}` (3 segments). ### License Proprietary — see LICENSE. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-gitops` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-gitops-azdevops` - List installed plugins: `vibe plugin list` ### Other GitOps plugins #### GitHub GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. #### GitLab GitLab provider for the GitOps meta plugin — REST v4. #### Bitbucket Cloud Bitbucket Cloud provider for the GitOps meta plugin — REST v2. --- # Bitbucket Cloud — VibeControls plugin > Bitbucket Cloud provider for the GitOps meta plugin — REST v2. Source: https://vibecontrols.com/plugins/gitops-bitbucket ## Bitbucket Cloud for VibeControls Bitbucket Cloud provider for the GitOps meta plugin — REST v2. ### About this plugin Bitbucket Cloud provider for the [`@vibecontrols/vibe-plugin-gitops`](https://npmjs.com/package/@vibecontrols/vibe-plugin-gitops) meta plugin. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-gitops # meta (once) vibe plugin install @vibecontrols/vibe-plugin-gitops-bitbucket ``` ### Auth App password with these permissions ([https://bitbucket.org/account/settings/app-passwords/](https://bitbucket.org/account/settings/app-passwords/)): - Repositories: read - Pull requests: read - Pipelines: read ``` vibe gitops auth set bitbucket "$BB_USERNAME:$BB_APP_PASSWORD" ``` PAT format: `username:app_password` (Basic auth). The username portion is your Bitbucket account name. ### Note Bitbucket Server / on-prem (Data Center) is NOT supported in v1 — only Bitbucket Cloud (bitbucket.org). ### License Proprietary — see LICENSE. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-gitops` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-gitops-bitbucket` - List installed plugins: `vibe plugin list` ### Other GitOps plugins #### GitHub GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. #### GitLab GitLab provider for the GitOps meta plugin — REST v4. #### Azure DevOps Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. --- # GitHub — VibeControls plugin > GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. Source: https://vibecontrols.com/plugins/gitops-github ## GitHub for VibeControls GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. ### About this plugin GitHub provider for the [`@vibecontrols/vibe-plugin-gitops`](https://npmjs.com/package/@vibecontrols/vibe-plugin-gitops) meta plugin. Talks to GitHub directly via REST v3 + GraphQL v4 — no `gh` CLI on the user box. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-gitops # meta (once) vibe plugin install @vibecontrols/vibe-plugin-gitops-github # this provider ``` ### Auth Personal Access Token (classic or fine-grained). Required scopes: - `repo` — repos, PRs, issues - `workflow` — Actions runs - `read:org` — org rollup - `security_events` — Dependabot / code scanning alerts ``` vibe gitops auth set github ghp_xxxxxxxxxxxxxxxxxxxx ``` ### Storage PAT lives in the agent's Skalex encrypted SQLite at namespace `gitops-github`, key `pat:default`. Plaintext token never leaves the agent. ### License Proprietary — see LICENSE. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-gitops` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-gitops-github` - List installed plugins: `vibe plugin list` ### Other GitOps plugins #### GitLab GitLab provider for the GitOps meta plugin — REST v4. #### Bitbucket Cloud Bitbucket Cloud provider for the GitOps meta plugin — REST v2. #### Azure DevOps Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. --- # GitLab — VibeControls plugin > GitLab provider for the GitOps meta plugin — REST v4. Source: https://vibecontrols.com/plugins/gitops-gitlab ## GitLab for VibeControls GitLab provider for the GitOps meta plugin — REST v4. ### About this plugin GitLab provider for the [`@vibecontrols/vibe-plugin-gitops`](https://npmjs.com/package/@vibecontrols/vibe-plugin-gitops) meta plugin. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-gitops # meta (once) vibe plugin install @vibecontrols/vibe-plugin-gitops-gitlab ``` ### Auth Personal Access Token (`api` scope). Self-hosted instances supported via `meta.host`: ``` vibe gitops auth set gitlab glpat-xxxxxxxxxxxxxxxxxxxx ``` Self-hosted GitLab: ``` curl -X POST "${AGENT_URL}/api/profiles/default/gitops/gitlab/auth" \ -H "x-agent-api-key: $KEY" \ -d '{"kind":"pat","token":"glpat-xxx","meta":{"host":"https://gitlab.acme.com"}}' ``` ### License Proprietary — see LICENSE. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-gitops` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-gitops-gitlab` - List installed plugins: `vibe plugin list` ### Other GitOps plugins #### GitHub GitHub provider for the GitOps meta plugin — REST v3 + GraphQL v4. #### Bitbucket Cloud Bitbucket Cloud provider for the GitOps meta plugin — REST v2. #### Azure DevOps Integrate Azure DevOps with VibeControls for seamless GitOps workflows. Supports REST 7.1 for efficient plugin management. --- # Plan plugins — VibeControls > Plan orchestration meta-plugin — owns the PlanProvider contract and dispatches to plan-mode providers like Plannotator. 1 provider available. Source: https://vibecontrols.com/plugins/plan ## Plan plugins for VibeControls Plan orchestration meta-plugin — owns the PlanProvider contract and dispatches to plan-mode providers like Plannotator. Structured AI planning providers — turn prompts into auditable step-by-step plans. ### 1 provider Each provider implements the Plan contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### Plannotator Plannotator plan provider — wraps the upstream CLI and configures per-AI-agent hooks. --- # Plannotator — VibeControls plugin > Plannotator plan provider — wraps the upstream CLI and configures per-AI-agent hooks. Source: https://vibecontrols.com/plugins/plan-plannotator ## Plannotator for VibeControls Plannotator plan provider — wraps the upstream CLI and configures per-AI-agent hooks. ### About this plugin Plannotator plan provider for the [VibeControls](https://vibecontrols.com) agent. Wraps the upstream [plannotator](https://plannotator.ai) CLI: - Installs the `plannotator` binary on demand (sha256-pinned). - Spawns a per-session plannotator HTTP server on a dedicated port. - Reverse-proxies the local UI at `/plan/:sessionId/*`, stripping iframe-blocking headers so the agent's vibetunnels URL can embed it. - Configures AI-agent hooks (Claude Code, OpenCode, Codex, Pi, Gemini) so completed plans land in plannotator automatically. - Registers with the agent's `ServiceRegistry` under the `"plan"` type so the meta plugin [`@vibecontrols/vibe-plugin-plan`](https://github.com/algoshred/vibe-plugin-plan) can dispatch to it. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-plan vibe plugin install @vibecontrols/vibe-plugin-plan-plannotator vibe plan-plannotator install # downloads the binary ``` Or set `VIBE_PLANNOTATOR_AUTOINSTALL=1` to auto-install on first plan request (not recommended for shared agents). ### REST API Mounted under `/api/plan-plannotator` on the agent (plus `/plan/:sessionId/*` as the reverse proxy). | Method | Path | Description | | --- | --- | --- | | `GET` | `/status` | Binary install status + running sessions | | `POST` | `/install` | Run the sha256-pinned installer | | `GET` | `/prereqs` | Same as `/status` (matches meta-plugin contract) | | `POST` | `/prereqs/install` | Alias of `/install` | | `POST` | `/stop?sessionId=` | Stop a specific session, or all sessions | | `GET` | `/agents/supported` | Detect installed AI agents + hook status | | `POST` | `/agents/:agent/configure-hook` | Write the plannotator hook for an agent | | `POST` | `/agents/:agent/unconfigure-hook` | Remove the hook | `:agent` is one of `claude`, `opencode`, `codex`, `pi`, `gemini`. ### CLI ``` vibe plan-plannotator status vibe plan-plannotator install vibe plan-plannotator stop [--session ] vibe plan-plannotator agents list vibe plan-plannotator agents configure vibe plan-plannotator agents unconfigure ``` ### Configuration Environment variables (all optional): | Name | Default | Meaning | | --- | --- | --- | | `PLANNOTATOR_PORT` | `19432` | Preferred port for plannotator. Falls back to a scan of +10. | | `VIBE_PLANNOTATOR_AUTOINSTALL` | unset | If `1`, auto-install the binary on first session start. | | `VIBE_PLANNOTATOR_INSTALL_UNPINNED` | unset | If `1`, bypass the sha256 pin on `install.sh` (NOT recommended). | | `VIBE_PLANNOTATOR_IDLE_MS` | `900000` | Kill a session after this many ms of inactivity. | | `VIBE_PLANNOTATOR_IDLE_SCAN_MS` | `60000` | Idle watchdog scan interval. | ### Per-agent hook setup - **Claude Code**: writes a `PreToolUse` hook matched to `ExitPlanMode` in `~/.claude/hooks.json`. - **OpenCode**: adds `@plannotator/opencode` to the `plugin` array in `~/.config/opencode/opencode.json`. - **Codex CLI**: appends a `[hooks.plannotator]` block bracketed by marker comments in `~/.codex/config.toml`. - **Pi**: drops `~/.pi/hooks/before_agent_start.d/vibe-plan-plannotator.sh`. - **Gemini CLI**: drops `~/.gemini/hooks/vibe-plan-plannotator.sh`. All hooks are idempotent (re-running configure does not duplicate) and refuse to write outside `$HOME`. ### Troubleshooting - **Port 19432 in use**: set `PLANNOTATOR_PORT` to another port. The provider scans the next 10 if the preferred port is busy. - **Iframe blocked**: confirm the agent's reverse proxy mounts cleanly by hitting `${tunnelUrl}/plan/` — expect a 404 placeholder, not a 502. - **Hung session**: `vibe plan-plannotator stop --session ` (or the 15 min idle watchdog will reap it). ### Development ``` bun install bun run sanity # format:check + lint + type:check + test + build ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-plan` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-plan-plannotator` - List installed plugins: `vibe plugin list` --- # Security plugins — VibeControls > Security lifecycle orchestrator — dispatches every dev / pr / merge / build / release stage to the right per-stage security provider. 14 providers available. Source: https://vibecontrols.com/plugins/security ## Security plugins for VibeControls Security lifecycle orchestrator — dispatches every dev / pr / merge / build / release stage to the right per-stage security provider. 14 per-lifecycle-stage security providers — secrets, SAST, DAST, SBOM, scorecards, signing, runtime, release gates. ### 14 providers Each provider implements the Security contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. #### Release gate OPA-backed release-gate at promote.prod — block on policy, evidence-required deploys. #### Scheduled rescan Nightly Grype offline rescan against last-built SBOMs. #### Runtime continuous Trivy + kube-bench continuous runtime checks against live workloads. #### SAST deep Full SAST + SCA sweep (Semgrep + osv-scanner) on pull_request.deep. #### SBOM build Syft CycloneDX SBOM + Grype scan at build lifecycle stage. #### OpenSSF Scorecard OpenSSF Scorecard checks at main.merge — track posture over time. #### Secrets (PR) Gitleaks-backed secret scanner for pull_request.fast — fail fast on PR. --- # Archive Offboard — VibeControls plugin > Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. Source: https://vibecontrols.com/plugins/security-archive ## Archive Offboard for VibeControls Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. ### About this plugin Archive / offboard provider for the `archive.offboard` lifecycle stage in [VibeControls](https://vibecontrols.com). Registers under provider name `tombstone-retention` against provider type `security.archive`, wrapping a pure-JS tombstone writer (`tombstone-retention@1.0.0`). **Wave 2 scaffold — real tool integration pending (except archive which is fully implemented for tombstones).** The host security meta plugin ([`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security)) dispatches archive runs to this provider when the user picks "tombstone-retention" as their default for `archive.offboard`. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-archive vibe security providers set-default --stage archive.offboard --provider tombstone-retention ``` No external binaries — pure JS. ### Behavior At `run()` the provider: - Reads the agent's local `security.sqlite` (via `host.getDataDir()`) and walks the `security_scan_runs` table for the vibe. Counts past scan runs and reads the most recent `conclusion`. Falls back to zero counts when the sqlite file is unavailable. - Builds a tombstone summary object containing: `vibeId`, `workspaceId`, `repoUrl` - `archivedAt` (ISO timestamp) - `scanRunCount`, `lastConclusion` - `retentionDays` (from `input.config`, default 90) - `producedBy` - Writes `tombstone.json` to `input.workdir`. - Surfaces the artifact in `evidence[]` so the host can upload it to long-term storage before clearing local rows. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["archive.offboard"].config`): ``` provider: tombstone-retention config: retentionDays: 90 # default; how long downstream archives should keep the tombstone ``` ### Evidence type note The `SecurityEvidenceType` union in `@vibecontrols/vibe-plugin-security` does not yet have a `"tombstone-json"` member. As a placeholder, the tombstone artifact is emitted with `type: "opa-decision"`. When the meta plugin adds `"tombstone-json"`, this provider will switch over (see TODO in `src/provider.ts`). Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-archive` - List installed plugins: `vibe plugin list` ### Other Security plugins #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # DAST (preview) — VibeControls plugin > OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. Source: https://vibecontrols.com/plugins/security-dast-preview ## DAST (preview) for VibeControls OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. ### About this plugin ### @vibecontrols/vibe-plugin-security-dast-preview OWASP ZAP baseline DAST provider for the `deploy.preview` lifecycle stage in [VibeControls](https://vibecontrols.com). Provider name: `zap-baseline`. Wraps the `ghcr.io/zaproxy/zaproxy:stable` Docker image and runs `zap-baseline.py` against the preview URL. Registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.release` (per `PROVIDER_TYPE_FOR_STAGE("deploy.preview")`). When the user picks "zap-baseline" as their default provider for `deploy.preview`, the security meta plugin dispatches scan runs here. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-dast-preview vibe security providers set-default --stage deploy.preview --provider zap-baseline ``` This plugin does NOT download a binary — it shells out to Docker. On `ensureToolInstalled()` the plugin probes `docker info` and best-effort `docker pull ghcr.io/zaproxy/zaproxy:stable` so the first real scan starts fast. If Docker is missing or unreachable, every scan returns a "succeeded" result with a single info finding describing the skip reason — the lifecycle never blocks on local-tooling absence. ### Behavior - **Probes** `docker info` (5s timeout). On failure, emits info finding `tool.docker.unavailable` with reason `docker-missing` and skips. - **HEAD-probes** `previewUrl` (10s timeout). On missing/failed probe, emits info finding `target.url.unreachable` with reason `target-url-missing` or `target-url-unreachable`. - **Runs**: @@MD@@``` docker run --rm -v :/zap/wrk/:rw zap-baseline.py \ -t -J report.json -r report.html -m 5 ```@@MD@@ Wall-clock budget controlled by `scanTimeoutSeconds` (default 600s). - **Parses** `report.json` into `NormalizedFinding[]`: Severity from ZAP `riskdesc`: `Critical → critical`, `High → high`, `Medium → medium`, `Low → low`, `Informational → info`. - Category: `sast` (closest fit; ZAP findings are HTTP/auth misconfigs). - `ruleId = zap.` (or kebab-cased name if no pluginid). - `cwe = ["CWE-"]` when present. - `file = `, `remediation = `. - **Emits** `report.json` + `report.html` as evidence artifacts. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["deploy.preview"].config`): ``` provider: zap-baseline config: previewUrl: https://pr-123.preview.example.com # required zapImage: ghcr.io/zaproxy/zaproxy:stable # optional override scanTimeoutSeconds: 600 # wall budget; default 600 extraArgs: [] # additional flags appended to zap-baseline.py ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-dast-preview` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # Deploy alpha smoke — VibeControls plugin > TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. Source: https://vibecontrols.com/plugins/security-deploy-alpha ## Deploy alpha smoke for VibeControls TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. ### About this plugin ### @vibecontrols/vibe-plugin-security-deploy-alpha Pure-JS smoke-check provider for the `deploy.alpha` lifecycle stage in [VibeControls](https://vibecontrols.com). Provider name: `alpha-smoke`. No binary, no Docker — uses the native `fetch` API only. **Wave 2 scaffold — real probe integration pending; see `src/provider.ts` TODO.** Registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.release` (per `PROVIDER_TYPE_FOR_STAGE("deploy.alpha")`). When the user picks "alpha-smoke" as their default provider for `deploy.alpha`, the security meta plugin dispatches scan runs here. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-deploy-alpha vibe security providers set-default --stage deploy.alpha --provider alpha-smoke ``` No external dependencies. Pure-JS, no subprocess, no privileged capabilities — runs anywhere the agent runs. ### Behavior (planned) For each URL in `input.config.alphaUrls`: - **TLS validity**: cert not expired, hostname matches, chain valid. - **Response headers**: `Strict-Transport-Security`, `X-Content-Type-Options`, `Content-Security-Policy` present on `GET /` (configurable per check). - **Auth challenge**: `GET /healthz` returns 200 OK with a known JSON shape for public endpoints; protected endpoints return 401 + `WWW-Authenticate`. Findings: - `severity: high` — TLS invalid or HSTS missing. - `severity: medium` — CSP or X-Content-Type-Options missing. - `severity: low` — `/healthz` auth-challenge mismatch. Evidence: raw probe results as a JSON artifact. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["deploy.alpha"].config`): ``` provider: alpha-smoke config: alphaUrls: - https://alpha.example.com - https://api.alpha.example.com requireHsts: true requireCsp: true requireXcto: true healthzPath: /healthz extraHeaders: User-Agent: vibecontrols-alpha-smoke ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-deploy-alpha` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # Developer local — VibeControls plugin > Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. Source: https://vibecontrols.com/plugins/security-developer-local ## Developer local for VibeControls Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. ### About this plugin `@vibecontrols/vibe-plugin-security-developer-local` serves the `developer.local` lifecycle stage. It registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.secrets` (which is shared with the `pull_request.*` secrets variants — see `PROVIDER_TYPE_FOR_STAGE` in the meta plugin) and the provider name `gitleaks-protect-semgrep`. It wraps `gitleaks protect` (uncommitted-only, fast) plus Semgrep `--quick` when Python is available on PATH, for a sub-second pre-commit experience. Wave 2 scaffold — real tool integration is pending; see `src/provider.ts` TODO. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-developer-local vibe security providers set-default --stage developer.local --provider gitleaks-protect-semgrep ``` The gitleaks binary is downloaded automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/gitleaks/`. Semgrep is best-effort and only invoked when `semgrep --version` resolves on PATH. ### Behavior (planned) - `gitleaks protect --staged --no-banner` against the working index — sub-second, never touches commit history. - Semgrep `--quick --config=p/ci` on changed files — only runs when `python3` + `semgrep` are on PATH. - Findings normalized to `category: "secret"` with `severity` derived from rule severity. - All evidence stored under the agent data dir; nothing leaves the host machine. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["developer.local"].config`): ``` provider: gitleaks-protect-semgrep config: semgrepEnabled: true # default; set false to skip Semgrep entirely extraGitleaksArgs: [] extraSemgrepArgs: [] ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-developer-local` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # Incident response — VibeControls plugin > Targeted incident-response scanner for incident.response lifecycle stage. Source: https://vibecontrols.com/plugins/security-incident ## Incident response for VibeControls Targeted incident-response scanner for incident.response lifecycle stage. ### About this plugin Targeted incident-response scanner for the `incident.response` lifecycle stage in [VibeControls](https://vibecontrols.com). Registers under provider name `cve-secret-blast-radius` against provider type `security.incident`, wrapping pinned Gitleaks (`8.21.2`) and Grype (`0.83.0`) restricted to the caller's targeted CVE / secret-rule subset. **Wave 2 scaffold — real tool integration pending.** The host security meta plugin ([`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security)) dispatches incident-triage scan runs to this provider when the user picks "cve-secret-blast-radius" as their default for `incident.response`. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-incident vibe security providers set-default --stage incident.response --provider cve-secret-blast-radius ``` The plugin downloads the Gitleaks + Grype binaries automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/{gitleaks,grype}/`. ### Planned behavior - Reads `input.config.cves[]` and runs `grype` restricted to those CVE IDs only (`--vuln-id` filter). Avoids the multi-minute baseline scan; sub-60s budget. - Reads `input.config.secretRules[]` and runs `gitleaks` with a synthesized `.gitleaks.toml` that whitelists only those rule IDs. - Merges output into a single `NormalizedFinding[]` (categories `vuln` + `secret`) and emits both SARIF (Gitleaks) + grype-json evidence so the incident artifact bundle is self-contained. - Hard timeout from `input.config.timeoutMs` (default 60 000) — partial results streamed on timeout. ### Configuration Per-vibe / per-incident config (stored in `RepositorySecurityConfig.pluginAssignments["incident.response"].config`): ``` provider: cve-secret-blast-radius config: cves: - CVE-2024-12345 - CVE-2024-67890 secretRules: - aws-access-token - generic-api-key timeoutMs: 60000 # hard cap for sub-60s triage ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-incident` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # Repo onboard — VibeControls plugin > Repo profile detector — picks the right policy bundle on repo.onboard. Source: https://vibecontrols.com/plugins/security-onboard ## Repo onboard for VibeControls Repo profile detector — picks the right policy bundle on repo.onboard. ### About this plugin `@vibecontrols/vibe-plugin-security-onboard` serves the `repo.onboard` lifecycle stage. It registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.onboard` and the provider name `repo-profile-detector`. It is a pure-JS detector — it does **not** wrap an external binary — and walks the repo root for marker files (`package.json`, `go.mod`, `Cargo.toml`, `Chart.yaml`, `manifest.json`, `pubspec.yaml`, `capacitor.config.*`, `main.tf`) to infer whether the repo is `frontend` / `backend` / `cli` / `mcp` / `chrome-extension` / `vscode-extension` / `mobile` / `iac` / etc. Wave 2 scaffold — real profile-classification model + policy seeding is pending; see `src/provider.ts` TODO. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-onboard vibe security providers set-default --stage repo.onboard --provider repo-profile-detector ``` ### Behavior - Walks `repoLocalPath` for high-precedence marker files in order: `capacitor.config.*`, `pubspec.yaml` → `mobile` - `manifest.json` → `chrome-extension` - `Chart.yaml`, `main.tf` → `iac` - `Cargo.toml` → `cli` - `go.mod`, `package.json` → `backend` - Emits a single info finding with `category: "config"`, `ruleId: "repo-profile-detector.profile-detected"`. - The detected profile + all matched markers are JSON-encoded in `rawProviderRef` for downstream policy seeders. ### Configuration No per-vibe configuration is required. The detector is idempotent and side-effect-free. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-onboard` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Package publish Cosign signing + SLSA provenance at package.publish. --- # Package publish — VibeControls plugin > Cosign signing + SLSA provenance at package.publish. Source: https://vibecontrols.com/plugins/security-package-publish ## Package publish for VibeControls Cosign signing + SLSA provenance at package.publish. ### About this plugin ### @vibecontrols/vibe-plugin-security-package-publish Cosign signing + SLSA provenance provider for the `package.publish` lifecycle stage in [VibeControls](https://vibecontrols.com). Provider name: `cosign-slsa`. Wraps the pinned `cosign` binary and emits a SLSA v1 in-toto provenance statement next to a cosign signature/certificate bundle. Registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.sbom` (per `PROVIDER_TYPE_FOR_STAGE("package.publish")`). When the user picks "cosign-slsa" as their default provider for `package.publish`, the security meta plugin dispatches scan runs here. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-package-publish vibe security providers set-default --stage package.publish --provider cosign-slsa ``` On first run the plugin auto-installs the pinned Cosign binary (sha256-verified per platform — linux-x64, linux-arm64, darwin-x64, darwin-arm64) into `~/.boff/vibecontrols/agents//tools/cosign/`. If `cosign` is already on `PATH` and reports the expected version, that binary is used as-is. ### Behavior - Validates `input.config.artifactPath`; concludes `skipped` (one info finding `signing.cosign.skipped`) if missing or unreadable. - Computes the sha256 of the artifact (used both for signing and the provenance subject digest). - **Cosign signing:** Key-based when `cosignKey` is provided (`COSIGN_PASSWORD` is taken from `cosignKeyPassword`). - Keyless via Fulcio OIDC otherwise — requires `SIGSTORE_OIDC_TOKEN`, `GITHUB_ACTIONS`, or `ACTIONS_ID_TOKEN_REQUEST_URL` in the environment. If absent, the cosign step is skipped with reason `keyless-oidc-unavailable` (provenance still runs). - On success, emits a `cosign-bundle` evidence artifact bundling the signature, certificate, artifact digest, and metadata. - **SLSA provenance:** Always runs (even when cosign is skipped) since the build identity + digest is independent of signing. - Writes a `provenance.intoto.jsonl` (in-toto Statement v1, default `predicateType = https://slsa.dev/provenance/v1`). - Subject: `{ name: , digest: { sha256: } }`. - Predicate: `buildDefinition.buildType = https://vibecontrols.com/build/v1`, builder id `https://vibecontrols.com/builder/vibe-plugin-security-package-publish`, run metadata (invocationId, startedOn, finishedOn). - Emits `provenance` evidence artifact. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["package.publish"].config`): ``` provider: cosign-slsa config: artifactPath: ./dist/app-v1.2.3.tar.gz # required cosignKey: /path/to/cosign.key # optional; omit for keyless cosignKeyPassword: "" # optional; sets COSIGN_PASSWORD fulcioUrl: https://fulcio.sigstore.dev # optional override rekorUrl: https://rekor.sigstore.dev # optional override slsaPredicateType: https://slsa.dev/provenance/v1 # optional extraCosignArgs: [] # extra flags appended before the artifact arg ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-package-publish` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # Release gate — VibeControls plugin > OPA-backed release-gate at promote.prod — block on policy, evidence-required deploys. Source: https://vibecontrols.com/plugins/security-release-gate ## Release gate for VibeControls OPA-backed release-gate at promote.prod — block on policy, evidence-required deploys. ### About this plugin OPA-backed release gate for the `promote.prod` lifecycle stage in [VibeControls](https://vibecontrols.com). Does **not** shell out — it calls the backend's `securityPolicyDecision(scanRunId)` GraphQL resolver, which in turn POSTs to the existing `opa-wspace:8181` ACA sidecar with the appropriate Rego bundle. Registers as a `security.release` provider (name `opa-release-gate`) with the [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) meta plugin. ### Decision input The backend resolver loads: - the scan run's `summary` (severity counts) - open `SecurityFinding[]` (joined to active `SecurityException[]`) - the workspace + per-vibe `SecurityPolicy` (level: `advisory` / `warn` / `block`) - artifact metadata (signed? SBOM present? pinned by digest?) It POSTs the document to `POST {OPA_URL}/v1/data/vibecontrols/security/release/allow` and caches the decision in Valkey for 5 minutes keyed by `(scanRunId, policyVersion)`. ### Provider output - `NormalizedFinding[]` — one row per `deny[]` entry with `category: "policy"` and severity derived from the policy expression - Evidence: a single `opa-decision` JSON document containing the input + decision (uploaded to S3 for the audit trail) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-release-gate` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # Scheduled rescan — VibeControls plugin > Nightly Grype offline rescan against last-built SBOMs. Source: https://vibecontrols.com/plugins/security-rescan ## Scheduled rescan for VibeControls Nightly Grype offline rescan against last-built SBOMs. ### About this plugin Scheduled nightly rescan provider for the `scheduled.rescan` lifecycle stage in [VibeControls](https://vibecontrols.com). Registers under provider name `grype-epss-rescan` against provider type `security.runtime`, wrapping pinned Grype (`0.83.0`) with optional EPSS score enrichment from FIRST.org. **Wave 2 scaffold — real tool integration pending.** The host security meta plugin ([`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security)) dispatches scan runs for `scheduled.rescan` to this provider when the user picks "grype-epss-rescan" as their default. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-rescan vibe security providers set-default --stage scheduled.rescan --provider grype-epss-rescan ``` The plugin downloads the Grype binary automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/grype/`. ### Planned behavior - Reload the latest SBOM evidence (cyclonedx-json) from the agent's local cache for the vibe being rescanned. - Run `grype sbom: -o json` in offline mode against the cached SBOM. - Normalize matches to `category: "vuln"` with severity derived from Grype's `severity` field. - For each finding with a CVE, optionally enrich with an EPSS score (`probability`, `percentile`) by calling `https://api.first.org/data/v1/epss?cve=`. ### Skip / fallback paths - **EPSS offline**: when the FIRST.org endpoint is unreachable (offline runners, blocked egress), the provider emits findings without EPSS scores and tags the evidence with `epss: "unavailable"`. - **Grype DB stale**: if the Grype vuln DB is older than the configured `maxDbAgeDays`, the provider records a `policy` severity-low finding noting the staleness and runs anyway. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["scheduled.rescan"].config`): ``` provider: grype-epss-rescan config: enrichWithEpss: true # call FIRST.org for each CVE epssEndpoint: https://api.first.org/data/v1/epss offline: false # set true to skip EPSS calls entirely maxDbAgeDays: 7 # warn if Grype DB is older than this ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-rescan` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # Runtime continuous — VibeControls plugin > Trivy + kube-bench continuous runtime checks against live workloads. Source: https://vibecontrols.com/plugins/security-runtime ## Runtime continuous for VibeControls Trivy + kube-bench continuous runtime checks against live workloads. ### About this plugin ### @vibecontrols/vibe-plugin-security-runtime Continuous runtime-check provider for the `runtime.continuous` lifecycle stage in [VibeControls](https://vibecontrols.com). Provider name: `trivy-kube-bench`. Wraps the pinned `trivy` binary and the `kube-bench` cluster check. **Wave 2 scaffold — real tool integration pending; see `src/provider.ts` TODO.** Registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.runtime` (per `PROVIDER_TYPE_FOR_STAGE("runtime.continuous")`). When the user picks "trivy-kube-bench" as their default provider for `runtime.continuous`, the security meta plugin dispatches scan runs here. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-runtime vibe security providers set-default --stage runtime.continuous --provider trivy-kube-bench ``` On first real run the plugin downloads the Trivy binary (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/trivy/`. `kube-bench` is invoked via `kubectl run` from a pinned image. ### Behavior (planned) - Rescans every image listed in `input.config.images` (or auto-discovered from the cluster) with Trivy (`--format sarif`), normalizing SARIF into `category: "vuln"` findings. - Runs kube-bench against the cluster control-plane / nodes (`--json`), normalizing the result into `category: "config"` findings. - **Skip path**: when no kubeconfig is present, the kubeconfig file is missing, or `kubectl get nodes` fails, the provider returns `succeeded` with a single info finding describing the skip reason (`missing-cluster-access`). The dispatcher records `conclusion: skipped`. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["runtime.continuous"].config`): ``` provider: trivy-kube-bench config: images: - registry.example.com/app:1.2.3 kubeconfig: /etc/vibe/kubeconfig # optional override registryAuth: server: registry.example.com username: vibe-readonly password: "" extraTrivyArgs: [] extraKubeBenchArgs: [] ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-runtime` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # SAST deep — VibeControls plugin > Full SAST + SCA sweep (Semgrep + osv-scanner) on pull_request.deep. Source: https://vibecontrols.com/plugins/security-sast-deep ## SAST deep for VibeControls Full SAST + SCA sweep (Semgrep + osv-scanner) on pull_request.deep. ### About this plugin `@vibecontrols/vibe-plugin-security-sast-deep` serves the `pull_request.deep` lifecycle stage. It registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.secrets` (which is shared with the other PR + developer.local secrets variants — see `PROVIDER_TYPE_FOR_STAGE` in the meta plugin) and the provider name `semgrep-osv-scanner`. It wraps Semgrep with the full ruleset (`--config auto`) for SAST and osv-scanner against package manifests for SCA. Wave 2 scaffold — real tool integration is pending; see `src/provider.ts` TODO. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-sast-deep vibe security providers set-default --stage pull_request.deep --provider semgrep-osv-scanner ``` The osv-scanner binary is downloaded automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/osv-scanner/`. Semgrep is downloaded for Linux only and falls back to PATH on darwin/windows (developers typically `pipx install semgrep`). ### Behavior (planned) - `semgrep scan --config auto --sarif --output /semgrep.sarif --metrics off` over `repoLocalPath` — full ruleset, SARIF normalized to `category: "sast"`. - `osv-scanner --recursive --format json --output /osv.json ` over package manifests (`package.json`, `go.mod`, `Cargo.toml`, `requirements.txt`, `Gemfile.lock`, `pom.xml`, etc.) — findings normalized to `category: "vuln"` with `cve` populated. - SARIF + osv-scanner JSON returned as evidence artifacts. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["pull_request.deep"].config`): ``` provider: semgrep-osv-scanner config: semgrepConfig: auto # or a comma-separated list of registry IDs / paths semgrepTimeoutSec: 600 osvIgnore: [] # CVE IDs to suppress extraSemgrepArgs: [] extraOsvArgs: [] ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-sast-deep` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # SBOM build — VibeControls plugin > Syft CycloneDX SBOM + Grype scan at build lifecycle stage. Source: https://vibecontrols.com/plugins/security-sbom-build ## SBOM build for VibeControls Syft CycloneDX SBOM + Grype scan at build lifecycle stage. ### About this plugin Generates a CycloneDX SBOM via Syft and scans it for known vulnerabilities via Grype. Targets the `build` lifecycle stage in [VibeControls](https://vibecontrols.com). Registers as a `security.sbom` provider (name `syft-grype`) with the [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) meta plugin. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-sbom-build vibe security providers set-default --stage build --provider syft-grype ``` Both Syft (Anchore) and Grype (Anchore) binaries are downloaded automatically on first use with sha256 verification. ### Outputs - `sbom.cdx.json` — CycloneDX 1.5 SBOM (evidence type `sbom-cyclonedx`) - `grype.json` — Grype vulnerability report (evidence type `grype-json`) - `NormalizedFinding[]` — one row per `matches[]` entry with CVE, package, fix version, severity Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-sbom-build` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # OpenSSF Scorecard — VibeControls plugin > OpenSSF Scorecard checks at main.merge — track posture over time. Source: https://vibecontrols.com/plugins/security-scorecard ## OpenSSF Scorecard for VibeControls OpenSSF Scorecard checks at main.merge — track posture over time. ### About this plugin `@vibecontrols/vibe-plugin-security-scorecard` serves the `main.merge` lifecycle stage. It registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.scorecard` and the provider name `openssf-scorecard`. It wraps the [OpenSSF Scorecard](https://github.com/ossf/scorecard) checks — running offline against a local clone for the file-only checks, and switching to the remote repo URL when `GH_TOKEN` is available on the agent host so the full check set runs (branch protection, signed commits, CI tests, etc.). Wave 2 scaffold — real tool integration is pending; see `src/provider.ts` TODO. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-scorecard vibe security providers set-default --stage main.merge --provider openssf-scorecard ``` The scorecard binary is downloaded automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/scorecard/`. ### Behavior (planned) Offline mode (no `GH_TOKEN`): - `scorecard --repo file:// --format json` — runs the file-only checks: Binary-Artifacts, License, Maintained, Pinned-Dependencies, Token-Permissions, CI-Tests (detected via `.github/workflows/*`). Online mode (`GH_TOKEN` set): - `scorecard --repo --format json` — runs the full check set: Branch-Protection, Signed-Releases, Code-Review, Vulnerabilities, Dependency-Update-Tool, SAST, Fuzzing, Security-Policy, Webhooks. Per-check findings are normalized to `category: "policy"` with severity derived from the score (0-3 = `high`, 4-6 = `medium`, 7-8 = `low`, 9-10 = `info`). The raw scorecard JSON is returned as evidence. ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["main.merge"].config`): ``` provider: openssf-scorecard config: checks: [] # subset of scorecard checks to run; empty = all online: auto # auto = use GH_TOKEN if present; force-offline / force-online overrides ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-scorecard` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # Secrets (PR) — VibeControls plugin > Gitleaks-backed secret scanner for pull_request.fast — fail fast on PR. Source: https://vibecontrols.com/plugins/security-secrets-pr ## Secrets (PR) for VibeControls Gitleaks-backed secret scanner for pull_request.fast — fail fast on PR. ### About this plugin Gitleaks-backed secret scanner for the `pull_request.fast` lifecycle stage in [VibeControls](https://vibecontrols.com). Registers itself with [`@vibecontrols/vibe-plugin-security`](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-security) under the per-stage provider type `security.secrets` and the provider name `gitleaks`. When the user picks "Gitleaks" as their default provider for `pull_request.fast` (or any other supported stage), the security meta plugin dispatches scan runs to this provider. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-security-secrets-pr vibe security providers set-default --stage pull_request.fast --provider gitleaks ``` The plugin downloads the Gitleaks binary automatically on first use (sha256-verified per platform) into `~/.boff/vibecontrols/agents//tools/gitleaks/`. ### Behavior - Output format: SARIF v2.1.0 via `--report-format=sarif --report-path=/gitleaks.sarif` - Scans the working tree at `repoLocalPath`; if `--depth` is configured, restricts to that many commits of history (default 200) - Findings normalized to `category: "secret"` with `severity` derived from `properties.severity` (Gitleaks rule field) when present, falling back to SARIF level - Secret samples redacted before persistence: only the sha256 + first-4 + last-4 chars are stored ### Configuration Per-vibe config (stored in `RepositorySecurityConfig.pluginAssignments["pull_request.fast"].config`): ``` provider: gitleaks config: configPath: .gitleaks.toml # optional, repo-local config file ignorePath: .gitleaksignore # optional, ignored matches re-reported as info historyDepth: 200 # commits of history to scan extraArgs: [] # additional CLI flags ``` Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-security` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-security-secrets-pr` - List installed plugins: `vibe plugin list` ### Other Security plugins #### Archive Offboard Writes tombstone.json evidence at archive.offboard for SOC2/ISO retention proofs. #### DAST (preview) OWASP ZAP baseline DAST scan against alpha preview URLs at deploy.preview. #### Deploy alpha smoke TLS / HSTS / CSP smoke checks for alpha environment URLs at deploy.alpha. #### Developer local Local pre-commit secrets scanner (gitleaks protect) — runs on the developer machine. #### Incident response Targeted incident-response scanner for incident.response lifecycle stage. #### Repo onboard Repo profile detector — picks the right policy bundle on repo.onboard. --- # Session Manager plugins — VibeControls > Unified session manager — capability discovery, feature negotiation, and provider routing across tmux, WezTerm, and Zellij. 3 providers available. Source: https://vibecontrols.com/plugins/session-manager ## Session Manager plugins for VibeControls Unified session manager — capability discovery, feature negotiation, and provider routing across tmux, WezTerm, and Zellij. Persistent remote tmux sessions, WezTerm panes, Zellij workspaces — all streamed to your browser. ### 3 providers Each provider implements the Session Manager contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### tmux Tmux + ttyd session provider — the default for persistent terminal sessions. #### WezTerm WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. #### Zellij Zellij + ttyd session provider — terminal workspace with floating panes. --- # tmux — VibeControls plugin > Tmux + ttyd session provider — the default for persistent terminal sessions. Source: https://vibecontrols.com/plugins/session-tmux ## tmux for VibeControls Tmux + ttyd session provider — the default for persistent terminal sessions. ### About this plugin Tmux + ttyd session provider plugin for [VibeControls Agent](https://www.npmjs.com/package/@vibecontrols/agent). ### Installation ``` vibe plugin install @vibecontrols/vibe-plugin-session-tmux ``` Or install globally alongside the agent: ``` npm install -g @vibecontrols/vibe-plugin-session-tmux ``` ### Features - **Tmux Sessions** -- Create, manage, and destroy tmux terminal sessions - **ttyd Integration** -- Web terminal access via ttyd (auto-started per session) - **Session Lifecycle** -- Full lifecycle management with health checks - **Command Execution** -- Run commands in existing tmux sessions - **Terminal Capture** -- Capture terminal output from running sessions - **Port Management** -- Automatic free port allocation for ttyd instances ### Provider Interface This plugin registers a `session` provider with the following capabilities: | Method | Description | | --- | --- | | `create(config)` | Create a new tmux session with optional ttyd terminal | | `get(id)` | Get session info by ID | | `list()` | List all managed sessions | | `terminate(id)` | Kill a tmux session and its ttyd process | | `execute(id, command)` | Send a command to a tmux session | | `capture(id)` | Capture current terminal output | | `startTerminal(id)` | Start a ttyd web terminal for a session | | `stopTerminal(id)` | Stop the ttyd process for a session | | `healthCheck()` | Check tmux and ttyd health | | `getSystemSessions()` | List all system tmux sessions | | `getSystemTerminals()` | List all running ttyd processes | ### Requirements - VibeControls Agent >= 2.0.0 - tmux installed on the host system - ttyd installed on the host system (optional, for web terminals) - Bun runtime >= 1.3.0 Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-session-manager` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-session-tmux` - List installed plugins: `vibe plugin list` ### Other Session Manager plugins #### WezTerm WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. #### Zellij Zellij + ttyd session provider — terminal workspace with floating panes. --- # WezTerm — VibeControls plugin > WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. Source: https://vibecontrols.com/plugins/session-wezterm ## WezTerm for VibeControls WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. ### About this plugin WezTerm + ttyd session provider plugin for [VibeControls Agent](https://www.npmjs.com/package/@vibecontrols/agent). Cross-platform alternative to the tmux session plugin — works on Windows, macOS, and Linux. ### Installation ``` vibe plugin install @vibecontrols/vibe-plugin-session-wezterm ``` Or install globally alongside the agent: ``` npm install -g @vibecontrols/vibe-plugin-session-wezterm ``` ### Features - **WezTerm Sessions** -- Create, manage, and destroy WezTerm terminal sessions via workspaces - **Cross-Platform** -- Works on Windows, macOS, and Linux (unlike tmux which is Unix-only) - **ttyd Integration** -- Web terminal access via ttyd (auto-started per session) - **Headless Operation** -- Uses `wezterm-mux-server` for headless session management - **Session Lifecycle** -- Full lifecycle management with health checks - **Command Execution** -- Run commands in existing WezTerm sessions - **Terminal Capture** -- Capture terminal output from running sessions - **Port Management** -- Automatic free port allocation for ttyd instances ### Provider Interface This plugin registers a `session` provider with the following capabilities: | Method | Description | | --- | --- | | `create(config)` | Create a new WezTerm session with optional ttyd terminal | | `get(id)` | Get session info by ID | | `list()` | List all managed sessions | | `terminate(id)` | Kill a WezTerm session and its ttyd process | | `execute(id, command)` | Send a command to a WezTerm session | | `capture(id)` | Capture current terminal output | | `startTerminal(id)` | Start a ttyd web terminal for a session | | `stopTerminal(id)` | Stop the ttyd process for a session | | `healthCheck()` | Check WezTerm and ttyd health | | `getSystemSessions()` | List all system WezTerm workspaces | | `getSystemTerminals()` | List all running ttyd processes | ### Requirements - VibeControls Agent >= 2.0.0 - WezTerm installed on the host system (with `wezterm` CLI in PATH) - ttyd installed on the host system (optional, for web terminals) - Bun runtime >= 1.3.0 Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-session-manager` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-session-wezterm` - List installed plugins: `vibe plugin list` ### Other Session Manager plugins #### tmux Tmux + ttyd session provider — the default for persistent terminal sessions. #### Zellij Zellij + ttyd session provider — terminal workspace with floating panes. --- # Zellij — VibeControls plugin > Zellij + ttyd session provider — terminal workspace with floating panes. Source: https://vibecontrols.com/plugins/session-zellij ## Zellij for VibeControls Zellij + ttyd session provider — terminal workspace with floating panes. ### About this plugin > Zellij + ttyd session provider — terminal workspace with floating panes. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-session-zellij ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-session-zellij # or npm install @vibecontrols/vibe-plugin-session-zellij ``` ### How it works Session **providers** implement the persistent-terminal contract from `@vibecontrols/vibe-plugin-session-manager` (meta). The meta plugin handles capability discovery, feature negotiation and per-session routing. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-session-manager` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **Zellij** — [https://github.com/zellij-org/zellij](https://github.com/zellij-org/zellij) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-session-zellij](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-session-zellij) - Source: [https://github.com/algoshred/vibe-plugin-session-zellij](https://github.com/algoshred/vibe-plugin-session-zellij) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/session-zellij](https://vibecontrols.com/plugins/session-zellij) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-session-manager` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-session-zellij` - List installed plugins: `vibe plugin list` ### Other Session Manager plugins #### tmux Tmux + ttyd session provider — the default for persistent terminal sessions. #### WezTerm WezTerm + ttyd session provider — GPU-accelerated multiplexer in the browser. --- # Storage plugins — VibeControls > Storage meta plugin — owns the AgentDatabase contract so embedded and shared backends are interchangeable behind one API. 2 providers available. Source: https://vibecontrols.com/plugins/storage ## Storage plugins for VibeControls Storage meta plugin — owns the AgentDatabase contract so embedded and shared backends are interchangeable behind one API. Pluggable agent storage. Embedded or shared backend — AES-256-GCM at rest either way. ### 2 providers Each provider implements the Storage contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### Skalex Document-oriented file-backed storage provider with AES-256-GCM at rest. #### Postgres Full-parity shared database storage provider with AES-256-GCM encryption — share state across multiple agents. --- # Postgres — VibeControls plugin > Full-parity shared database storage provider with AES-256-GCM encryption — share state across multiple agents. Source: https://vibecontrols.com/plugins/storage-postgres ## Postgres for VibeControls Full-parity shared database storage provider with AES-256-GCM encryption — share state across multiple agents. ### About this plugin > Full-parity PostgreSQL storage provider with AES-256-GCM encryption — share state across multiple agents. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-storage-postgres ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-storage-postgres # or npm install @vibecontrols/vibe-plugin-storage-postgres ``` ### How it works Storage **providers** implement the `AgentDatabase` contract from `@vibecontrols/vibe-plugin-storage` (meta) so the agent's persistence layer is pluggable across embedded and server backends. This package is a **provider** registered against the `@vibecontrols/vibe-plugin-storage` meta plugin. Install the meta plugin first; this provider plugs into it. ### Upstream - **PostgreSQL** — [https://www.postgresql.org/](https://www.postgresql.org/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-storage-postgres](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-storage-postgres) - Source: [https://github.com/algoshred/vibe-plugin-storage-postgres](https://github.com/algoshred/vibe-plugin-storage-postgres) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/storage-postgres](https://vibecontrols.com/plugins/storage-postgres) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-storage` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-storage-postgres` - List installed plugins: `vibe plugin list` ### Other Storage plugins #### Skalex Document-oriented file-backed storage provider with AES-256-GCM at rest. --- # Skalex — VibeControls plugin > Document-oriented file-backed storage provider with AES-256-GCM at rest. Source: https://vibecontrols.com/plugins/storage-skalex ## Skalex for VibeControls Document-oriented file-backed storage provider with AES-256-GCM at rest. ### About this plugin Skalex storage provider for the VibeControls agent. Document-oriented, file-backed, encrypted at rest via AES-256-GCM. Zero native dependencies — runs natively on Bun. ### Usage Importing this package registers the adapter under the name `"skalex"` with `@vibecontrols/vibe-plugin-storage`. The agent picks it up automatically: ``` import "@vibecontrols/vibe-plugin-storage-skalex"; // registers "skalex" import { createAgentDatabase } from "@vibecontrols/vibe-plugin-storage"; const db = await createAgentDatabase({ dbPath: "/path/to/data", encryptionKey: "f".repeat(64), // 32 bytes hex }); ``` ### Encryption All on-disk data is AES-256-GCM encrypted via Skalex's built-in `encrypt: { key }` option. Each collection is gzipped + ciphertext on disk. The end-to-end test (`scripts/e2e-encryption.sh`) bootstraps the full chain and grep-asserts the absence of plaintext on disk. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-storage` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-storage-skalex` - List installed plugins: `vibe plugin list` ### Other Storage plugins #### Postgres Full-parity shared database storage provider with AES-256-GCM encryption — share state across multiple agents. --- # code-server — VibeControls plugin > Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. Source: https://vibecontrols.com/plugins/tool-code-server ## code-server for VibeControls Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. ### About this plugin > Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-tool-code-server ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-tool-code-server # or npm install @vibecontrols/vibe-plugin-tool-code-server ``` ### How it works **Tool** plugins surface a single capability on the agent's REST surface and (where applicable) an iframe-mounted UI. They run independently of any meta plugin. ### Upstream - **code-server** — [https://github.com/coder/code-server](https://github.com/coder/code-server) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-code-server](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-code-server) - Source: [https://github.com/algoshred/vibe-plugin-tool-code-server](https://github.com/algoshred/vibe-plugin-tool-code-server) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/tool-code-server](https://vibecontrols.com/plugins/tool-code-server) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tool-code-server` - List installed plugins: `vibe plugin list` ### Related Tool plugins #### Git client Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. #### GraphiQL API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. #### SSH SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. --- # Git client — VibeControls plugin > Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. Source: https://vibecontrols.com/plugins/tool-git ## Git client for VibeControls Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. ### About this plugin > Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-tool-git ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-tool-git # or npm install @vibecontrols/vibe-plugin-tool-git ``` ### How it works **Tool** plugins surface a single capability on the agent's REST surface and (where applicable) an iframe-mounted UI. They run independently of any meta plugin. ### Upstream - **Git** — [https://git-scm.com/](https://git-scm.com/) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-git](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-git) - Source: [https://github.com/algoshred/vibe-plugin-tool-git](https://github.com/algoshred/vibe-plugin-tool-git) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/tool-git](https://vibecontrols.com/plugins/tool-git) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tool-git` - List installed plugins: `vibe plugin list` ### Related Tool plugins #### code-server Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. #### GraphiQL API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. #### SSH SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. --- # GraphiQL — VibeControls plugin > API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. Source: https://vibecontrols.com/plugins/tool-graphiql ## GraphiQL for VibeControls API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. ### About this plugin > GraphQL Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. ### Install ``` vibe plugin install @vibecontrols/vibe-plugin-tool-graphiql ``` Or install the npm package directly into an existing project that hosts the VibeControls agent: ``` bun add @vibecontrols/vibe-plugin-tool-graphiql # or npm install @vibecontrols/vibe-plugin-tool-graphiql ``` ### How it works **Tool** plugins surface a single capability on the agent's REST surface and (where applicable) an iframe-mounted UI. They run independently of any meta plugin. ### Upstream - **GraphiQL** — [https://github.com/graphql/graphiql](https://github.com/graphql/graphiql) ### More - npm: [https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-graphiql](https://www.npmjs.com/package/@vibecontrols/vibe-plugin-tool-graphiql) - Source: [https://github.com/algoshred/vibe-plugin-tool-graphiql](https://github.com/algoshred/vibe-plugin-tool-graphiql) - Plugin contract / SDK: [https://github.com/algoshred/vibecontrols-plugin-sdk](https://github.com/algoshred/vibecontrols-plugin-sdk) - Plugin catalogue: [https://vibecontrols.com/plugins/tool-graphiql](https://vibecontrols.com/plugins/tool-graphiql) Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tool-graphiql` - List installed plugins: `vibe plugin list` ### Related Tool plugins #### code-server Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. #### Git client Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. #### SSH SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. --- # SSH — VibeControls plugin > SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. Source: https://vibecontrols.com/plugins/tool-ssh ## SSH for VibeControls SSH connections, remote terminals, port forwarding, and remote agent installation — all from the dashboard. ### About this plugin SSH connections & port forwarding plugin for [VibeControls Agent](https://www.npmjs.com/package/@vibecontrols/agent). ### Platform support This plugin is **POSIX-only** for now (Linux, macOS, WSL). It shells out to `ssh`, `scp`, `chmod`, `tar`, and uses `nohup` to launch `ttyd` on the remote host. Native Windows support (cmd / PowerShell) is not yet implemented — the plugin self-disables on `process.platform === "win32"` and the CLI subcommands exit early with a "not supported on Windows yet" message. Track progress in the issue tracker. ### Installation ``` vibe plugin install @vibecontrols/vibe-plugin-ssh ``` Or install globally alongside the agent: ``` npm install -g @vibecontrols/vibe-plugin-ssh ``` Then register it: ``` vibe plugin list # verify it shows up ``` ### Features - **SSH Connections** — Save, test, and manage SSH connection configs - **Remote Execution** — Execute commands on remote servers via SSH - **Port Forwarding** — Create and manage SSH port forwards (local → remote) - **Real-time Output** — SSH command output streamed via Socket.IO events ### API Routes Once installed, the plugin registers these routes on the agent: | Method | Path | Description | | --- | --- | --- | | GET | `/api/ssh/connections` | List all SSH connections | | POST | `/api/ssh/connections` | Create a new SSH connection | | POST | `/api/ssh/execute` | Execute a remote command | | POST | `/api/ssh/test/:id` | Test an SSH connection | | DELETE | `/api/ssh/connections/:id` | Delete a connection | | GET | `/api/port-forward/` | List all port forwards | | POST | `/api/port-forward/` | Create a port forward | | POST | `/api/port-forward/:id/start` | Start forwarding | | POST | `/api/port-forward/:id/stop` | Stop forwarding | | DELETE | `/api/port-forward/:id` | Delete a forward | ### CLI Commands SSH and port forward CLI commands are built into the `vibe` CLI: ``` vibe ssh list # List saved connections vibe ssh add --name my-server ... # Add a connection vibe ssh test -i # Test connectivity vibe ssh exec -i -c "uptime" # Run remote command vibe forward list # List port forwards vibe forward create --local 5432 --remote-host db --remote-port 5432 --server my-server vibe forward start -i # Start forwarding vibe forward stop -i # Stop forwarding ``` ### Requirements - VibeControls Agent >= 1.1.0 - Node.js >= 18.0.0 Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tool-ssh` - List installed plugins: `vibe plugin list` ### Related Tool plugins #### code-server Browser-based VS Code via code-server — reverse-proxied through the VibeControls agent. #### Git client Visual Git client (Ungit) — reverse-proxied through the VibeControls agent. #### GraphiQL API Playground (GraphiQL) — embedded in the VibeControls agent for any endpoint. --- # Tunnel plugins — VibeControls > Unified tunnel manager — routes requests across registered tunnel providers (Cloudflare, VibeTunnels). 2 providers available. Source: https://vibecontrols.com/plugins/tunnel ## Tunnel plugins for VibeControls Unified tunnel manager — routes requests across registered tunnel providers (Cloudflare, VibeTunnels). Secure localhost tunneling. Cloudflare-backed for public exposure or frp-based for air-gapped relay. ### 2 providers Each provider implements the Tunnel contract for a specific backend. Install the meta plugin plus the provider(s) you need. #### Cloudflare Tunnel Cloudflare Tunnel provider — expose localhost endpoints via cloudflared with named tunnels. #### VibeTunnels frp-based tunnel provider — self-hostable tunnel relay for air-gapped networks. --- # Cloudflare Tunnel — VibeControls plugin > Cloudflare Tunnel provider — expose localhost endpoints via cloudflared with named tunnels. Source: https://vibecontrols.com/plugins/tunnel-cloudflare ## Cloudflare Tunnel for VibeControls Cloudflare Tunnel provider — expose localhost endpoints via cloudflared with named tunnels. ### About this plugin Cloudflare Tunnel provider for [VibeControls Agent](https://www.npmjs.com/package/@vibecontrols/agent). ### Installation ``` vibe plugin install @vibecontrols/vibe-plugin-tunnel-cloudflare ``` Or install globally alongside the agent: ``` npm install -g @vibecontrols/vibe-plugin-tunnel-cloudflare ``` ### Features - **Cloudflare Tunnels** -- Create and manage Cloudflare quick tunnels (no account required) - **Auto-Start** -- Automatically starts an agent tunnel on server boot - **URL Extraction** -- Extracts the tunnel URL from cloudflared output - **Process Management** -- Graceful start/stop with process cleanup - **Storage Persistence** -- Tunnel state persisted via agent KV storage ### Provider Interface This plugin registers a `tunnel` provider with the following capabilities: | Method | Description | | --- | --- | | `start(config)` | Start a new Cloudflare tunnel for a local port | | `stop(id)` | Stop a running tunnel | | `get(id)` | Get tunnel info by ID | | `list()` | List all managed tunnels | | `delete(id)` | Delete a tunnel record | | `getStatus()` | Get overall tunnel provider status | ### Requirements - VibeControls Agent >= 2.0.0 - `cloudflared` installed on the host system - Bun runtime >= 1.3.0 Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-tunnel` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tunnel-cloudflare` - List installed plugins: `vibe plugin list` ### Other Tunnel plugins #### VibeTunnels frp-based tunnel provider — self-hostable tunnel relay for air-gapped networks. --- # VibeTunnels — VibeControls plugin > frp-based tunnel provider — self-hostable tunnel relay for air-gapped networks. Source: https://vibecontrols.com/plugins/tunnel-vibetunnels ## VibeTunnels for VibeControls frp-based tunnel provider — self-hostable tunnel relay for air-gapped networks. ### About this plugin VibeTunnels frp-based tunnel provider for the VibeControls Agent. Spawns `frpc` to expose local ports via a shared `frps` server running in the VibeTunnels EKS clusters, producing public URLs at `*.vibetunnels.com` (prod) / `*.alpha.vibetunnels.com` (alpha). Registers itself as a `tunnel` provider named `tunnel-vibetunnels` inside the agent's service registry. Works together with `@vibecontrols/vibe-plugin-tunnel` (the manager/facade plugin that owns the `/api/tunnels/*` REST surface). ### Install ``` bun install -g @vibecontrols/vibe-plugin-tunnel-vibetunnels ``` Installed automatically by `vibe start` when VibeTunnels is the default tunnel provider. ### Requirements - `frpc` binary in `$PATH`, OR set `VIBETUNNELS_FRPC_PATH` to an absolute path, OR let the plugin download a pinned build on first use (cached under `~/.boff/vibecontrols/cache/frpc/`). ### How it works - The backend issues a tunnel session via GraphQL `issueTunnelSession`, returning a `controlPlanePayload` containing the frps server address, auth token, proxy name, and managed hostname. - The CLI POSTs that payload to the local agent's `/api/tunnels/issue-session`, which routes through the tunnel manager plugin to this provider. - The provider writes an `frpc.toml` config into `~/.boff/vibecontrols/agents/{profile}/plugins/tunnel-vibetunnels/{tunnelId}.toml` and persists a TunnelInfo record in encrypted agent storage. - `start(tunnelId)` spawns `frpc -c ` and scrapes the subprocess output for `start proxy success` to transition the tunnel to `active`. - `stop(tunnelId)` gracefully terminates the frpc subprocess. - `rotate(tunnelId)` issues a new session and re-spawns frpc — this causes a ~1s interruption because frp OSS does not support dynamic config reload. ### Security - frpc auth tokens are never logged. - All storage goes through the agent's encrypted Skalex adapter. - The plugin verifies the frpc binary checksum against a pinned manifest before execution when auto-downloading. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install the meta first: `vibe plugin install @vibecontrols/vibe-plugin-tunnel` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-tunnel-vibetunnels` - List installed plugins: `vibe plugin list` ### Other Tunnel plugins #### Cloudflare Tunnel Cloudflare Tunnel provider — expose localhost endpoints via cloudflared with named tunnels. --- # UI Hello — VibeControls plugin > Minimal example UI plugin — copy as a starting point for new iframe-mounted plugin UIs. Source: https://vibecontrols.com/plugins/ui-hello ## UI Hello for VibeControls Minimal example UI plugin — copy as a starting point for new iframe-mounted plugin UIs. ### About this plugin Sample UI plugin used by the VibeControls doctor's Playwright E2E spec. Contributes to **both** `vibe.detailTab` and `agent.detailTab` from the same package so the cross-mount-point story is exercised end-to-end. ### Layout ``` src/ index.ts # Plugin manifest (VibePluginFactory) ui-dist/ index.html # Iframe content (hand-rolled v1 envelope) ``` The static UI is dependency-free on purpose — when the doctor test fails, the iframe payload is easy to read. ### Build ``` bun install bun run build # bundles src/index.ts → dist/index.js ``` The `ui-dist/` directory is shipped as-is alongside `dist/` and served by the agent at `/ui/vibe-plugin-ui-hello`. ### Install on a running agent ``` vibe plugin install @vibecontrols/vibe-plugin-ui-hello ``` Then open `https://vibecontrols.local.burdenoff.com/vibecontrols/vibes/` with the `vibes.tabs.iframe-host` feature flag enabled — the "Hello" tab appears in the strip. Use it in 30 seconds - Make sure the agent is running: `vibe start` - Install this plugin: `vibe plugin install @vibecontrols/vibe-plugin-ui-hello` - List installed plugins: `vibe plugin list` --- # Newsroom & Media Kit - VibeControls > VibeControls press releases, official logos, banners, brochures, and brand guidelines. Media resources and press contact. Source: https://vibecontrols.com/press ## Newsroom & Media Kit VibeControls press releases, official logos, brand assets, and media resources — everything you need to write about us. ### Logos #### VibeControls Icon #### VibeControls Logo #### VibeControls Wordmark ### Documents #### Partner / Reseller Agreement The full VibeControls Partner / Reseller Agreement — review the program terms before applying. #### VibeControls Partner Cheat Sheet One page: what VibeControls is, who buys it, what it costs, and how to introduce us. #### VibeControls Partner Playbook The full brief: positioning, buyer personas, use cases, pricing, competitive landscape, objection handling, lead handoff and partner commercials — plus every product screen. #### Burdenoff Product Portfolio Every Burdenoff product on one sheet, with the signals that tell you which one a conversation is pointing at. ### Media Contact For interviews, partnerships, and press inquiries: [contact@vibecontrols.com](mailto:contact@vibecontrols.com) --- # Pricing Plans - VibeControls > Simple, transparent pricing for VibeControls. Choose Individual or Enterprise plans. Source: https://vibecontrols.com/pricing ## Simple, Transparent Pricing Choose the plan that's right for you. No hidden fees, no surprises. ### Pricing Start free on SaaS. Scale through Hobby, Pro, and Team. Enterprise unlocks dedicated tenancy and self-hosting. #### VibeControls - Free Choose this plan - ✓ 1 agent - ✓ 1 workspace · 5 vibes - ✓ 3 active sessions (2h limit) - ✓ 1 SSH target · 1 tunnel - ✓ 5 AI sessions/day (BYOK) - ✓ 500 MB storage - ✓ Community support #### VibeControls - Hobby (Monthly) Choose this plan - ✓ 5 agents - ✓ 1 workspace · 10 vibes - ✓ 10 active sessions (4h limit) - ✓ 7 plugin installs - ✓ 25 AI sessions/day (BYOK) - ✓ 5 GB storage - ✓ Email support #### VibeControls - Pro (Monthly) Choose this plan - ✓ 20 agents - ✓ 3 workspaces · 100 vibes - ✓ Unlimited sessions - ✓ 50 SSH targets · 25 tunnels - ✓ 100 AI sessions/day (BYOK) - ✓ 25 GB storage - ✓ Priority support (<24h) - ✓ API access --- # Security - VibeControls > Enterprise-grade security at VibeControls. Source: https://vibecontrols.com/security ## Security Last updated: May 16, 2026 ### Enterprise-Grade Security Security is built into every layer of VibeControls. ### Data Encryption All data encrypted in transit (TLS 1.3) and at rest (AES-256). Credentials stored in HSMs. ### Authentication & Access Control - Multi-factor authentication (MFA) - SAML/SSO integration - Role-based access control (RBAC) - API key management with scoped permissions ### Infrastructure Security - Kubernetes with network policies - Istio service mesh with mTLS - Private networks, no public DB access - DDoS protection via Cloudflare ### Compliance - Comprehensive audit logging - GDPR-compliant data handling - Regular third-party security audits ### Report a Vulnerability Contact security@vibecontrols.com for responsible disclosure. --- # VibeControls Expert Services > Leverage expert services for VibeControls: agent deployment, custom plugins, DevOps automation, secure tunneling, and more. Source: https://vibecontrols.com/services ## Expert hands on VibeControls VibeControls is a product you can run yourself. But you do not have to do it alone. We help you **adopt and get more from** VibeControls — and we deliver **done-for-you outcomes powered by it**: agent fleets, custom plugins, AI-native workflows, governance, and enablement. Onboarding, setup, consulting, migration, security, and training so your team gets the full value of the platform — faster, and with governance from day one. Done-for-you outcomes we deliver using the product — custom plugins, DevOps automation, and managed agent fleets — so you can focus on building. ### What we do Comprehensive services across the VibeControls lifecycle — every one grounded in a real product capability. Engage one, or combine several into a program. #### Agent Deployment & Environment Setup We stand up VibeControls agents across your fleet — local laptops, remote SSH hosts, containers, and cloud VMs — and wire each target into a clean, observable topology from day one. - DIRECT (localhost) and SSH target design, with SSH host auto-discovery - Remote agent install from an existing agent — no manual binary uploads - Multi-instance and multi-environment topology (local · staging · prod · client servers) - Per-agent JSON config, health monitoring, and Agent Graph reachability review - Encrypted agent-local state with the right storage provider (embedded or shared) #### Custom Plugin & VibeModule Development Need a capability that is not in the marketplace yet? We design and build it as a first-class plugin against the three-tier model so it loads, upgrades, and audits like everything else. - New session, tunnel, AI, or storage providers behind the meta-plugin interfaces - Standalone tool plugins with their own REST routes, UI panels, and CLI commands - Vibe Deck actions, webhook handlers, and template-bundled automation - Prerequisite declarations and safe auto-install hooks - Sandbox-tested builds and a clear path to publish in your workspace or the marketplace #### DevOps & Infrastructure Automation We turn your day-two operations into one-click, scheduled, and audited workflows on the agent — replacing the brittle pile of scripts, cron jobs, and tribal knowledge. - Vibe Deck buttons for deploy, restart, health-check, and runbook steps - Vibe Calendar scheduled tasks with logs, history, and retries - Inbound/outbound webhooks for CI/CD, GitHub, and Slack with test + replay - Reproducible build snapshots and file-system export jobs into the Files service - Quota-aware automation that stays inside RBAC and audit from the first run #### Agent Orchestration & AI-Native Workflow Consulting We help your team design AI-native workflows that actually ship — provider routing, plan-then-execute safety, and human-in-the-loop checkpoints on high-blast-radius actions. - Per-vibe AI contexts, prompt libraries, and BYOK provider strategy (16+ providers) - Plan mode + Plannotator review gates for migrations and mass refactors - AI Assistant governed-action design via delegated app-as-user grants - AI Builder enablement to generate plugins and actions from natural language - Sandbox-backed execution so generated code runs isolated before touching the project #### Secure Tunneling & Access Setup We configure governed tunnels so local and remote services are reachable through workspace-owned URLs — not random links pasted into chat — with metering and expiry built in. - Cloudflare and VibeTunnels providers, shards, domains, and per-hour usage metering - Port-forwarding to browsable URLs and raw-TCP (tcpmux) for SSH, databases, non-HTTP - Browser-based code-server access tunneled through your agent - Expiry, ownership, and quota policy aligned to your security posture - Demo and preview environments exposed safely for stakeholders and beta testers #### Managed Agents & Fleet Operations Prefer to focus on building? We operate your agent fleet for you — monitoring health, rotating credentials, keeping plugins current, and acting on what the topology tells us. - Continuous health and last-seen monitoring across every target - Agent lifecycle: start, stop, restart, force-kill, and version upgrades - Per-agent plugin inventory upkeep — install, enable, disable, health-check - Scheduled maintenance, backups, and nightly task queues - Quota, usage, and audit reporting surfaced back to your team #### Migration & Onboarding from Existing Tooling Moving off a tangle of tmux + SSH + wikis, or another control plane? We migrate your environments, projects, and runbooks into vibes with minimal disruption. - Inventory of current environments, scripts, schedules, and access paths - Modelling projects as vibes — path, git remote, env vars, tags, and notes - Porting runbooks into Vibe Deck actions and Vibe Calendar schedules - Per-vibe documentation sites and notes migrated and published - Cutover plan, parallel-run validation, and team handover #### Security Hardening, RBAC & Governance We tune VibeControls to your compliance bar — least-privilege roles, audited actions, encryption choices, and security plugins that ride along with the developer workflow. - Workspace RBAC role design and least-privilege access reviews - BYOK at-rest encryption configuration for repo, sandbox, and secret material - Security plugins: release gate, SBOM build, and PR secret scanning - Scorecard baseline (Bronze/Silver/Gold) and gated-action policy design - Audit-log review, retention configuration, and evidence workflows #### Training & Enablement for Engineering Teams We get your engineers genuinely productive — not just logged in. Role-based sessions, golden-path templates, and onboarding flows so a new hire is shipping in hours, not weeks. - Hands-on workshops for developers, DevOps, and platform teams - Golden-path and vibe/sandbox templates tailored to your stack - Plugin SDK and AI Builder enablement so teams extend the platform themselves - Onboarding checklists, tours, and curated learning paths - CLI, MCP, VS Code, and SDK adoption to meet developers where they work #### IDP & Software Catalog Enablement Turn VibeControls into your Internal Developer Platform: a live software catalog fed by what developers actually run, with ownership, templates, and self-service that stays in sync with reality. - Catalog component ↔ vibe binding so the catalog reflects what is really running - Typed component, ownership, and lifecycle metadata setup - Template gallery and self-service action design for golden paths - Dependency and topology views for impact analysis - Adoption metrics and catalog-health review as the rollout progresses #### We build it with you — custom plugins, actions, and integrations Our engineers work alongside yours to extend VibeControls exactly to your workflow. Every capability ships as a proper plugin against the three-tier model, tested in a sandbox before it touches your project, so it upgrades and audits like everything else. - New providers behind the session, tunnel, AI, and storage interfaces - Vibe Deck actions, webhooks, and template-bundled automation - Sandbox-tested builds with a clean path to publish #### Or hand us the pager — managed agents and fleet operations When you would rather focus on building, we run the control plane for you. We monitor agent health across every environment, keep plugins current, rotate credentials, and act on what the topology tells us — with quota, usage, and audit reporting back to your team. - Continuous health monitoring and lifecycle management - Scheduled maintenance, backups, and nightly task queues - RBAC-scoped, fully audited operations — never shared secrets ### A clear path from intent to impact The same loop whether it is a one-week setup or a multi-quarter platform program — scoped, reviewable, and audited at every step. #### Discover We map your environments, teams, tooling, and goals — and agree on what "done" looks like before any work starts. #### Plan A scoped, reviewable plan: target topology, plugins, RBAC model, and rollout sequence — with human-in-the-loop checkpoints where it matters. #### Build We deploy agents, develop plugins, wire automation, and validate everything in sandboxes — shipping in increments you can see. #### Operate We hand over with docs and enablement, or stay on to run the fleet — monitored, audited, and quota-aware from the first day. #### We leave your team able to run it without us Every engagement ships documentation, golden-path templates, and hands-on enablement. Whether you keep us on to operate the fleet or take the keys yourself, your engineers understand the platform — not just the dashboard. - Role-based workshops for developers, DevOps, and platform teams - Plugin SDK and AI Builder enablement so the team extends it themselves - Onboarding checklists, tours, and curated learning paths ### Let's scope what you need Tell us about your environments and goals. We will recommend the right mix of setup, development, operations, and enablement — and a plan to get there. Prefer a structured partnership? [See our partner program](https://vibecontrols.com/partners). --- # Store – VibeControls > Browse all vibeplugins, integrations, and templates for VibeControls. Source: https://vibecontrols.com/store ## VibeControls Store Discover vibeplugins, integrations, and templates built for the VibeControls developer control platform. ### Most Popular Top downloads across the community #### Git Tool Visual Git client (Ungit) reverse-proxied through your VibeControls agent — browse history, stage, commit, and push from the agent UI. ### Built something useful? Share your vibeplugins, integrations, and templates with the VibeControls community. Earn from your work or contribute for free. By submitting a listing you agree to the [Marketplace & Developer Agreement](https://burdenoff.com/developer-agreement). --- # Browse VibeControls Store | VibeControls plugin > Browse all vibeplugins, integrations, and templates for VibeControls. Source: https://vibecontrols.com/store/browse ## Browse the VibeControls Store Browse all vibeplugins, integrations, and templates for VibeControls. Type Pricing Category Available types - PLUGIN 1 #### Git Tool Visual Git client (Ungit) reverse-proxied through your VibeControls agent — browse history, stage, commit, and push from the agent UI. --- # Use Cases — VibeControls > 45 real workflows showing how engineers, platform teams, and enterprises use VibeControls — the Agentic Engineering IDP for AI-Native teams. Source: https://vibecontrols.com/use-cases ## What can you build with VibeControls? Real use cases grounded in the product — from solo developer setups to enterprise governance workflows. Pick your persona and find your story. Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. Individual Developer ### Instant public URL for your local dev server Share your local dev server via a public HTTPS URL instantly. No VPN or staging needed—just two clicks to share with your team or clients. Individual Developer ### Shared AI prompt library with project context Create a shared AI prompt library with reusable project context, variables, categories, tags, and workspace sharing. Individual Developer ### Send code to AI directly from VS Code Send code to AI directly from VS Code with VibeControls. Highlight, right-click, and get AI responses tailored to your active Vibe.'s project context automatically attached. Individual Developer ### Give your AI assistant scoped access to your codebase Generate scoped URLs for AI access to specific code files. Keep your full codebase private while providing necessary context to AI. DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. DevOps / SRE ### One-click incident response operations deck Streamline incident response with VibeDeck. Execute common actions like API restarts and cache flushes with a single click. DevOps / SRE ### Self-serve operations for the whole org via shared VibeDeck Enable self-serve operations with shared VibeDeck. Product teams can deploy, flush caches, and perform health checks without SSH access. DevOps / SRE ### Connect your database GUI to a remote database over a tunnel Connect your database GUI to a remote database using VibeTunnels. No VPN or IP allow-listing needed for seamless access. DevOps / SRE ### Instant Slack alert when a session fails Get instant Slack alerts for session failures. Subscribe to session.error events and notify your team without polling or dashboards. DevOps / SRE ### Queue AI test-generation jobs and review results in the morning Queue AI test jobs overnight with VibeControls. Review generated test cases in the morning, ready to integrate into your test suite. DevOps / SRE ### Roll out new capabilities to your whole fleet from the plugin catalog Manage plugins across your fleet with VibeControls. Browse, install, and update plugins without SSH, directly from the catalog. DevOps / SRE ### The entire platform from your terminal — and your CI pipeline Access the full VibeControls platform via CLI. Automate tasks with scriptable output and CI-ready authentication. Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. Team & Platform ### Living software catalog with real ownership and dependency graphs Maintain a live software catalog with VibeControls. Register services, APIs, and libraries with ownership and dependency graphs.'s connected to the Vibes where work actually happens. Team & Platform ### Auto-built docs site embedded in your project Vibe Embed an auto-built docs site in your project Vibe. Use Markdown for documentation and eliminate the need for separate platforms. Team & Platform ### See exactly what version is deployed in every environment Monitor deployed versions across environments with the Environment Topology view. Ensure safe promotions with a full audit trail. Team & Platform ### Pair on a live terminal session from a Slack link Share terminal sessions via Slack with teammates. Collaborators can join read-only or request control. All actions are audited. Team & Platform ### Quality scorecards your CI keeps honest Create scorecards for vibes and components. Metrics update with each build, showing pass/warn/fail status and trend history. Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. Enterprise & Governance ### Connect AI coding tools to an agent-local plan workspace Set up the agent-local Plannotator integration for supported AI CLIs, then use its embedded workspace to inspect the plan sessions those tools create. Enterprise & Governance ### Fine-grained RBAC without complexity Define workspace roles with RBAC. Roles map to access needs, providing project-level control without added complexity. AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. AI-Native & Extensibility ### Give external beta testers time-limited API access via tunnel token Provide beta testers with time-limited API access. Create a public-access tunnel with a 4-hour expiry for testing. AI-Native & Extensibility ### A workspace AI assistant that acts as you — never beyond you Use AI Assistant for workspace actions. It operates under your identity, respecting RBAC permissions and auditing. AI-Native & Extensibility ### One source of truth for your AGENTS.md and .cursorrules Centralize AGENTS.md and .cursorrules in VibeControls. Scope instructions to agents or entire Vibes, ensuring consistent team coding standards. AI-Native & Extensibility ### See exactly what your AI agents did — task by task Track AI agent tasks in real-time with VibeControls. View task status, source tool, and agent details in a live stream for full transparency. Enterprise & Governance ### Scan every vibe and agent — then triage the findings in one queue Conduct comprehensive security scans on Vibes and agents. Manage findings in a unified queue, set statuses, and document exceptions efficiently. Team & Platform ### Package any operation into a reusable, shareable action Create reusable Actions from commands, scripts, or webhooks. Share and run pre-built Actions with parameters across your team seamlessly. AI-Native & Extensibility ### Run Claude Code, Codex, Gemini and OpenCode from one workbench Integrate Claude, Codex, Gemini, and more in one workbench. Control AI sessions with model, SDK mode, and server settings in a single interface. AI-Native & Extensibility ### Bring your own model keys — and keep them on your own machine Securely store API keys on your machine. VibeControls supports various providers without needing to access your keys for AI harness sessions. AI-Native & Extensibility ### Decide how much rope the agent gets — per session, per machine Customize agent autonomy per session. Choose from Plan, Accept edits, or Full auto modes, ensuring local command origin for security. AI-Native & Extensibility ### AI spend you can see — checked before the turn, charged after it Monitor AI credit usage with VibeControls. Check budgets before dispatch and charge actual token counts post-turn for accurate spend tracking.'s own prices, deduplicated by message id, and shown in a meter inside the panel. AI-Native & Extensibility ### Turn what you just debugged into reusable AI context Capture and reuse AI contexts from terminal sessions. Share prompts across teams and integrate them into harness chats or workbench drawers. AI-Native & Extensibility ### AI that proposes the ops fix and waits for your approval VibeControls previews AI-driven ops plans for agent issues. Human approval required for execution, with governance features already active. ### Ready to start your own story? Free plan. One agent, every feature, BYOK AI — start in under a minute. --- # Package any operation into a reusable, shareable action — VibeControls Use Cases > Create reusable Actions from commands, scripts, or webhooks. Share and run pre-built Actions with parameters across your team seamlessly. Source: https://vibecontrols.com/use-cases/action-store-library ## Package any operation into a reusable, shareable action Author parameterized actions once, install ready-made ones from the store, and run them with inputs across your team VibeDecks put one-click ops buttons on a board, but the real primitive underneath is the Action — a named, reusable operation you author once and run anywhere. Most teams never realize they can work with actions directly, so the same "restart the service" or "rotate the key" script gets rewritten in five slightly different places. VibeControls makes Actions a first-class library. Open the Actions page and create one: pick a type, drop in the command or a script (with your chosen interpreter), set the environment variables it needs, and decide whether it requires a confirmation before it runs. Save it and it is now a durable, named building block — not a snippet buried in someone's shell history. You do not always have to start from scratch. Install a ready-made action straight from the store by its store item and version, optionally renaming it for your workspace. It lands in your library as a real Action you can inspect, run, and adapt. Found one that is close but not exact? Duplicate it and tweak the copy without touching the original. Running an action is parameterized: pass inputs at execution time so a single "deploy" action serves every environment instead of ten near-identical copies. Each run is recorded as an execution you can review, and a long-running one can be cancelled if it needs to be stopped. Sharing spreads the leverage. Share an action with teammates or mint a share link, all governed by the same workspace RBAC as every other resource, and bind it to a VibeDeck button when you want a one-click surface on top. Author once, install from the store, parameterize, share — the reusable action library is how a platform team stops rewriting the same operation and starts curating a catalog of them. ### Do it yourself Author a reusable Action with typed inputs, install ready-made ones from the store, run them with parameters, and share them across the team. - 1 Open the Actions page /vibecontrols/actions Go to Actions to see your reusable action library. You should see: You see the actions available in your workspace. [Open in app](https://app.vibecontrols.com/vibecontrols/actions) Mark done - 2 Author a new action /vibecontrols/actions Click New, choose the type, add the command or a script with its interpreter, set environment variables, and decide whether it needs a confirmation before running. - 3 Install a ready-made action from the store /vibecontrols/actions Browse the store, pick an action by item and version, optionally rename it, and install. - 4 Run it with inputs /vibecontrols/actions Execute the action and pass parameters at run time so one action serves every environment. - 5 Share it or wire it to a VibeDeck button /vibecontrols/actions Share the action with teammates or mint a share link, and bind it to a VibeDeck button for a one-click surface. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Scan every vibe and agent — then triage the findings in one queue Run Claude Code, Codex, Gemini and OpenCode from one workbench ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # One source of truth for your AGENTS.md and .cursorrules — VibeControls Use Cases > Centralize AGENTS.md and .cursorrules in VibeControls. Scope instructions to agents or entire Vibes, ensuring consistent team coding standards. Source: https://vibecontrols.com/use-cases/agent-instruction-files ## One source of truth for your AGENTS.md and .cursorrules Author, version, and share the instruction files that steer every coding agent — no more drifting CLAUDE.md copies The quality of an AI coding agent is only as good as the instructions steering it. Yet those instructions — AGENTS.md, CLAUDE.md, GEMINI.md, a hand-tuned .cursorrules — usually live as loose files copied from machine to machine, drifting out of sync the moment one person edits their copy. Two engineers on the same project end up with two subtly different sets of house rules, and their agents behave differently. VibeControls turns instruction files into managed, first-class entities. Open the Instructions tab on a Vibe, click New, and pick the filename type — AGENTS_MD, CLAUDE_MD, GEMINI_MD, or CUSTOM for anything else (name it ".cursorrules" and that is exactly what lands on disk). Paste in your standards — architecture notes, coding conventions, the commands the agent may and may not run — and save. The content is stored centrally, and its real on-disk filename is derived automatically from the type you chose. Scope decides reach. An AGENT-scoped file applies to a single machine; a VIBE-scoped file follows the whole project, so every agent that picks up that Vibe writes the same rules to its working directory. Change the canonical file once and the update propagates instead of stranding stale copies on individual laptops. Sharing is built in. Grant a teammate access to an instruction file, or mint a share link, and they inherit the exact same steering — governed by the same workspace RBAC that protects every other resource. Edits are attributed and timestamped, so you always know who changed the house rules and when. The result is a single source of truth for how your agents should behave. Instead of hoping everyone remembered to update their local CLAUDE.md, you author the instructions once in VibeControls, scope them, share them, and let every coding agent — Claude Code, Cursor, Gemini, or your own — read from the same page. ### Do it yourself Author an AGENTS.md or .cursorrules once inside VibeControls, scope it to a Vibe so every agent writes the same rules to its working directory, and share it with the team. - 1 Open the Instructions tab on your Vibe /vibecontrols/vibes Go to Vibes, open the project Vibe you want to steer, and switch to the Instructions tab. You should see: You see the instruction files currently attached to this Vibe (empty on a new Vibe). [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Create a new instruction file /vibecontrols/vibes Click New and pick the filename type — AGENTS.md, CLAUDE.md, GEMINI.md, or Custom (name it ".cursorrules" for a Cursor rules file). - 3 Write your house rules /vibecontrols/vibes Paste in your conventions — architecture notes, coding standards, allowed and forbidden commands — and save. - 4 Set the scope /vibecontrols/vibes Choose the scope: VIBE so it follows the whole project across every agent, or AGENT to target a single machine. - 5 Share it with the team /vibecontrols/vibes Open the share control on the file, grant a teammate access or mint a share link, and set their role. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling A workspace AI assistant that acts as you — never beyond you See exactly what your AI agents did — task by task ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Visual map of your entire infrastructure topology — VibeControls Use Cases > Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. Source: https://vibecontrols.com/use-cases/agent-topology-graph ## Visual map of your entire infrastructure topology See which machines can reach which — at a glance When you're managing machines across a VPC, a home network, a cloud provider, and a contractor's laptop, "can A reach B?" becomes a daily debugging question. Most teams answer it by SSHing around and hoping for the best. VibeControls' Agent Graph gives you a persistent topology view. The graph plots every registered target and agent as a node, with directed edges for the mesh connections you have configured between them. Node labels show the saved target and agent status, while edge selection exposes the connection permissions and scopes. You can see which agents have a configured path to your staging database versus which do not. From a node, you can open a mesh request — VibeControls validates the reachable root and every authorized hop before proxying the requested command or API call to the destination. For incident response, the topology view is the first place to inspect your configured relationships and the latest displayed node status. Pairwise latency probes and automatic live connection-status updates are not currently controls in this graph. ### Do it yourself Map registered targets, agents, and configured mesh connections, adjust connection permissions, and send a validated request through an authorized mesh path. - 1 Open the Agent Graph /vibecontrols/agents?tab=graph Go to the Agents page and click the Graph tab. You should see: The mesh topology graph renders, showing your infrastructure at a glance. [Open in app](https://app.vibecontrols.com/vibecontrols/agents?tab=graph) Mark done - 2 See your agents and targets as nodes /vibecontrols/agents?tab=graph Look over the graph canvas where every registered agent and target appears as a node. - 3 Trace which agents can reach which /vibecontrols/agents?tab=graph Follow the directed edges between nodes to see the peer connections each agent has across the fleet. - 4 Run a request through the mesh /vibecontrols/agents?tab=graph Click a node, open the Mesh Request panel, pick a destination, and send a command or API call through the mesh. - 5 Refresh the topology as the fleet changes /vibecontrols/agents?tab=graph Return to the graph after adding agents, targets, or mesh connections, and use the displayed node status to inspect the saved topology. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Bootstrap a multi-server agent fleet in under 5 minutes Centrally review scheduled database-backup runs across servers ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. DevOps / SRE ### One-click incident response operations deck Streamline incident response with VibeDeck. Execute common actions like API restarts and cache flushes with a single click. --- # A workspace AI assistant that acts as you — never beyond you — VibeControls Use Cases > Use AI Assistant for workspace actions. It operates under your identity, respecting RBAC permissions and auditing. Source: https://vibecontrols.com/use-cases/ai-assistant-governed-actions ## A workspace AI assistant that acts as you — never beyond you Page-aware chat, delegated actions under your identity, metered by AI credits AI assistants in dev tools usually fall into one of two traps: a chatbot that can describe your problem but can't touch anything, or an over-permissioned bot wired to a god-mode service account that does whatever anyone asks. Neither belongs near production tooling. The VibeControls AI Assistant takes a third path. It's a slide-over panel available on every page of the app, and it knows where you are — open it on an agent's detail page and ask "why is this agent unhealthy?" and the question carries that page context automatically. Each conversation runs in its own isolated sandbox provisioned on workspace compute, so there's nothing to install — no agent required on your machine. Modes select what the assistant can do. In general mode it answers questions about your environment. Switch to api-calls mode and it can act — restart a session, inspect your vibes, fix what you approve — through delegated authentication that executes every operation under your identity. If your role can't perform an action, neither can the assistant: the same RBAC that governs you returns FORBIDDEN to it. Every mutation it performs shows up in the audit log attributed to you. Usage is metered by AI credits, visible in a meter right in the panel. Credits are checked before each message and token usage is reported after, so consumption is transparent — and running out produces a clear out-of-credits state with guidance, never a silent failure. Multi-provider fallback across Anthropic, OpenAI, and OpenRouter keeps the assistant answering through provider hiccups. Long conversations survive — the sandbox runtime extends rather than resetting your context — and the useful ones outlive the chat: bookmark key messages or export the whole conversation for the team. ### Do it yourself Open the always-available AI Assistant on any page, let it take real actions as you within your permissions, track your AI credits, and bookmark or export the conversation. - 1 Open the AI Assistant /vibecontrols On any page in the app, press Ctrl+Shift+A or click the AI Assistant toggle to slide the panel open. You should see: A slide-over assistant panel appears, already aware of the page you're on. [Open in app](https://app.vibecontrols.com/vibecontrols) Mark done - 2 Check your AI credits /vibecontrols Glance at the credit meter in the panel header to see how many AI credits you have available. - 3 Ask a question in General mode /vibecontrols Use the mode selector to pick General, then type your question and send it. - 4 Switch to API Calls mode and ask it to act /vibecontrols Choose API Calls in the mode selector, then ask the assistant to perform a real task on your behalf. - 5 Stay within your permissions /vibecontrols Ask the assistant to do something you don't have access to, and watch how it responds. - 6 Keep the conversation going /vibecontrols Continue the conversation with follow-up messages over a long session. - 7 Bookmark a key message /vibecontrols/ai/bookmarks Open the message menu (the ⋯) on an important reply and choose Bookmark. - 8 Export the conversation /vibecontrols Open the export menu in the panel and pick JSON or Markdown. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Give external beta testers time-limited API access via tunnel token One source of truth for your AGENTS.md and .cursorrules ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Decide how much rope the agent gets — per session, per machine — VibeControls Use Cases > Customize agent autonomy per session. Choose from Plan, Accept edits, or Full auto modes, ensuring local command origin for security. Source: https://vibecontrols.com/use-cases/ai-autonomy-dial ## Decide how much rope the agent gets — per session, per machine Plan, accept-edits or full-auto, over an agent that refuses remote execution by default "Let the agent run commands" is a one-way door in most tools: a single global setting, flipped once, that quietly applies to every future session on every machine. That is a bad shape for a decision this consequential. VibeControls makes autonomy a per-session dial with three real settings. Plan is read-only — the agent reasons and proposes, it does not edit or execute. Accept edits applies file changes automatically but still asks before risky commands. Full auto lets it run commands and edit files without asking, and selecting it requires an explicit confirmation rather than a silent toggle. The dial is provider-agnostic and only offered for CLI-mode sessions; SDK sessions are chat-only, so there is nothing to grant. Underneath the dial sits a second, stricter layer that lives on the agent itself. The agent default-denies shell, script and file write/delete tasks whenever the request did not come from localhost — so a perfectly valid workspace token driving your machine over a tunnel still cannot execute a command until you opt that specific agent in. Two named gate keys control it: one for shell and script execution, one for file mutation, and the file gate stays off even when you enable shell. When a run is blocked you get a stable error code and an in-context dialog offering to enable the gate, not a mysterious failure. For harnesses that support it there is a third integration surface: Plan mode. VibeControls installs the Plannotator provider on the agent and walks you through configuring a hook for Claude, OpenCode, Codex, Gemini, or Pi. The selected CLI and provider then decide when a plan is created and how its review affects execution; VibeControls supplies the agent connection, hook setup, and embedded provider workspace rather than a platform-wide execution gate. The three layers compose. A brand-new session can explore read-only on a machine that has never been opted in; a branch you trust can run accept-edits; full auto can be reserved for one dedicated agent whose gate keys you deliberately set. Because the gates are agent configuration, they are per-machine and change under the same permissions and audit trail as any other agent change — which means "who let the AI run shell here, and when" is a question with an answer. ### Do it yourself Set per-session autonomy, keep the agent refusing remote execution until you opt it in, and put Plan mode in front of anything that runs. - 1 Start a CLI-mode session /vibecontrols/ai In the AI workbench, open a session on your agent and switch the SDK selector to CLI mode so the autonomy control becomes available. You should see: The autonomy selector appears in the composer for that session. [Open in app](https://app.vibecontrols.com/vibecontrols/ai) Mark done - 2 Choose the autonomy level /vibecontrols/ai Pick Plan for read-only reasoning, Accept edits to auto-apply file changes but still be asked about risky commands, or Full auto — which asks you to confirm explicitly. - 3 Check the machine-level gate /vibecontrols/agents/:id?tab=security Open the agent's Security tab and look at the remote task permissions. By default the agent refuses shell, script and file-mutation requests that did not originate locally. - 4 See a blocked run explain itself /vibecontrols/actions Run a command action against an agent whose gate is still off, from the Actions page or a VibeDeck button. - 5 Put Plan mode in front of the agent /vibecontrols/ai-tools/plan Open Plan mode, install the plan plugins and CLI if prompted, then run the wizard to configure the hook for Claude, OpenCode, Codex, Gemini or Pi. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Bring your own model keys — and keep them on your own machine AI spend you can see — checked before the turn, charged after it ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Build a custom Vibe plugin by describing it in plain language — VibeControls Use Cases > Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. Source: https://vibecontrols.com/use-cases/ai-builder-plugin ## Build a custom Vibe plugin by describing it in plain language AI Builder scaffolds the plugin, UI, handlers, and sandbox test environment The VibeControls plugin SDK is powerful, but writing a plugin from scratch still requires understanding the manifest format, lifecycle hooks, REST route conventions, and UI panel integration. Most developers who could benefit from a custom plugin never build one. AI Builder lowers the barrier to zero. Open the AI Builder interface and describe what you want: "a VibeDeck-style panel with a button that runs our Alembic database migration script on the deployment target, captures stdout and stderr, and displays pass/fail status with the migration names that ran." The AI Builder generates the complete plugin scaffold: the VibePlugin manifest, the action handler, the REST route, and the UI panel. It also creates a sandbox test environment — an isolated execution context where you can run the plugin against a safe replica of your environment before installing it on real agents. Review the scaffold, edit what needs changing, and click deploy. The plugin appears in the workspace's plugin catalog and can be installed on any agent with one click. What used to require a plugin developer and a sprint now requires a platform engineer and a sentence. Teams that could never afford bespoke tooling now build it themselves. ### Do it yourself Describe a capability in plain language, generate a complete plugin scaffold, test it in a safe sandbox, and publish it to your workspace plugin catalog in one click. - 1 Open the AI Builder /vibecontrols/ai-builder From the navigation, open AI Builder. This is the dedicated space for creating new Vibe plugins. You should see: The AI Builder opens with a simple describe form, ready for you to define your plugin. [Open in app](https://app.vibecontrols.com/vibecontrols/ai-builder) Mark done - 2 Describe your plugin in plain language /vibecontrols/ai-builder Give your plugin a name, pick the kind of capability it adds (terminal, AI, tunnel, integration, or UI), and write what you want it to do in plain words. - 3 Generate the plugin scaffold /vibecontrols/ai-builder Click Generate scaffold. The builder produces a complete, valid plugin skeleton for you. - 4 Test it in an isolated sandbox /sandbox Send the handler to the sandbox Script Runner to try it out in a safe, isolated environment. - 5 Refine with the AI Assistant /vibecontrols/ai-builder Hand the scaffold off to the Vibe Plugins assistant to iterate conversationally and polish the logic. - 6 Publish to your workspace catalog /vibecontrols/ai-builder Click Publish to catalog. Your finished plugin is added to the workspace Component Catalog in one click. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Control VibeControls from Claude Code with 600+ tools Review AI-generated scripts in an isolated sandbox before real execution ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. AI-Native & Extensibility ### Give external beta testers time-limited API access via tunnel token Provide beta testers with time-limited API access. Create a public-access tunnel with a 4-hour expiry for testing. --- # AI spend you can see — checked before the turn, charged after it — VibeControls Use Cases > Monitor AI credit usage with VibeControls. Check budgets before dispatch and charge actual token counts post-turn for accurate spend tracking. Source: https://vibecontrols.com/use-cases/ai-credit-metering ## AI spend you can see — checked before the turn, charged after it A pre-flight credit gate, a cost-weighted charge per completed turn, no silent failures AI features usually hide their cost until the invoice arrives, and they fail in the worst possible way when a budget runs out: mid-answer, with no explanation. VibeControls wraps every assistant turn in two explicit steps. Before the message is dispatched, a pre-flight credit check runs; if the monthly budget is depleted the gateway rejects it and the panel blocks the send and shows you the upgrade path, rather than starting an answer it cannot finish. After the turn completes, the real token counts — prompt and completion — are reported and charged. What is charged is computed on the server from the model's own token prices, never from a number the browser supplies, and the charge is keyed on the assistant message it belongs to. A retried or duplicated report therefore cannot double-bill you. Because the weighting is per model, a cheap model costs proportionally less than an expensive one instead of every turn costing the same flat amount. What you see is a credit meter in the assistant panel that refreshes after each turn, and an explicit out-of-credits state with guidance when the budget is gone. Harness sessions have their own view: the per-agent AI stats tab reports session and usage counts straight from the agent, because those turns spend your own provider key and are not metered in credits at all. Two honest notes. First, that split is the whole point — platform inference is metered, your own keys are not, and there is no path where VibeControls quietly bills you for tokens you already paid a provider for. Second, workspace-admin cost dashboards and anomalous-spend alerts are on the roadmap, not shipped: today the transparency is per user and per agent, not an org-wide spend view. ### Do it yourself Watch the assistant pre-check your AI credits before each message, charge the real cost-weighted token usage afterwards, and see harness-session usage separately per agent. - 1 Open the AI Assistant /vibecontrols On any page, open the AI Assistant panel. You should see: The panel slides over with a credit meter in view. [Open in app](https://app.vibecontrols.com/vibecontrols) Mark done - 2 Read the credit meter /vibecontrols Check the meter before you start a long conversation. - 3 Send a message and watch the pre-flight check /vibecontrols Ask the assistant something. The budget is checked before the message is dispatched. - 4 See the charge land after the turn /vibecontrols Let the reply finish and glance at the meter again. - 5 Check harness-session usage per agent /vibecontrols/ai-tools/ai/stats Open the AI stats view and pick an agent. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Decide how much rope the agent gets — per session, per machine Turn what you just debugged into reusable AI context ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Run Claude Code, Codex, Gemini and OpenCode from one workbench — VibeControls Use Cases > Integrate Claude, Codex, Gemini, and more in one workbench. Control AI sessions with model, SDK mode, and server settings in a single interface. Source: https://vibecontrols.com/use-cases/ai-harness-workbench ## Run Claude Code, Codex, Gemini and OpenCode from one workbench Every coding agent on every machine you own, in one tabbed console Every coding agent is its own island. Claude Code keeps its config in ~/.claude, OpenCode in ~/.config/opencode, Goose somewhere else again. Each has its own auth, its own session history, its own idea of where "the project" is. Run three of them across two machines and you are holding six mental models at once — and none of them can see the others' work. VibeControls treats a harness as a plugin on the agent rather than a tool you install by hand. Install the AI orchestrator (@vibecontrols/vibe-plugin-ai) on a registered agent, then add provider plugins for the harnesses you want — the picker offers eight that are validated end to end: Claude, Codex, Gemini, OpenCode, Cursor, OpenRouter, Minimax and Ollama. The moment the first provider is detected the AI workbench lights up; nothing reloads, nothing needs restarting. The workbench is a three-pane console: sessions on the left, the active conversation in the middle, session detail on the right, and a tab bar across the top so several live sessions on several agents stay open at once. Sessions can be renamed, searched, bulk-deleted or terminated, and keyboard shortcuts move between tabs without touching the mouse. Because every session belongs to a real agent, the chat is talking to your actual machine — your repo, your toolchain, your network. Each turn is configurable rather than fixed. Pick the model from the list the provider reports on that agent. Switch between SDK mode (chat-only) and CLI mode. Choose which MCP servers this session may reach. Point the working directory at the agent root or at a specific Vibe's path. Attach files, dictate with voice input, and set the autonomy level for the turn. These are session controls, not global settings, so a careful review session and an aggressive refactor session can run side by side. Switching harness normally means re-teaching it your house rules. The Bridle panel manages the cross-harness configuration files and skills — Claude Code, OpenCode, Goose, Amp, Copilot and Crush — from one place, and instruction files scoped to a Vibe land in every agent's working directory. The harness changes; the standards do not. ### Do it yourself Install the AI orchestrator and the harnesses you use on an agent, then run Claude, Codex, Gemini or OpenCode sessions side by side with per-turn control over model, mode and MCP servers. - 1 Install the AI plugins on an agent /vibecontrols/agents/:id Open your agent, go to the Plugins tab, and install the AI orchestrator plus the provider plugins for the harnesses you want (Claude, Codex, Gemini, OpenCode, Cursor, OpenRouter, Minimax or Ollama). You should see: The agent reports the AI plugins as installed and the workbench unlocks for it. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 Save the provider key /vibecontrols/ai In the AI workbench, open the AI Keys panel from the admin bar and save the API key for the provider you just installed. - 3 Open the AI workbench /vibecontrols/ai Go to the AI workbench and pick the agent you want to work on. - 4 Start a session and set the turn controls /vibecontrols/ai Create a session, then use the composer controls to choose the model, the SDK or CLI mode, which MCP servers it may reach, and whether the working directory is the agent root or a Vibe path. - 5 Run several harnesses side by side /vibecontrols/ai Open more sessions — on the same agent or a different one — and switch between them with the tab bar. - 6 Keep the harness config consistent /vibecontrols/ai/config Open the AI configuration view and use the Bridle panel to manage profiles and skills across Claude Code, OpenCode, Goose, Amp, Copilot and Crush. - 7 Save the replies worth keeping /vibecontrols/ai/bookmarks Bookmark an important message from a session, then open your saved messages page. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Package any operation into a reusable, shareable action Bring your own model keys — and keep them on your own machine ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # AI that proposes the ops fix and waits for your approval — VibeControls Use Cases > VibeControls previews AI-driven ops plans for agent issues. Human approval required for execution, with governance features already active. Source: https://vibecontrols.com/use-cases/ai-ops-decision-plans ## AI that proposes the ops fix and waits for your approval Roadmap preview: operational proposals that wait for human approval before any tool fires This is a roadmap preview, not a walkthrough you can run today. Today every operational decision in VibeControls is a human reading a dashboard. An agent goes degraded and somebody notices the card. A sandbox burns quota past its TTL over a weekend and somebody eventually spots it. A catalog component drifts from what is actually deployed and nobody finds out until an incident. The platform has the data; it just does not act on it. The pieces that make acting safely possible are already shipped, and they are the hard parts. Plan mode is a real human-in-the-loop gate that turns a proposal into something you read, annotate, and approve or deny before any tool call fires. The AI tool-event ledger already records what AI work happened, with a live stream into the app. Scheduled prompt dispatch already exists as a primitive for sending work to an agent session on a timetable. AI credits already provide a pre-flight budget check. And the AI workbench is already the surface those things live on. What is designed and not yet built is the decision layer: a small set of proposal operations that reason over the state VibeControls already holds and emit a plan rather than a mutation. Agent remediation for a machine that is degraded or stuck awaiting configuration. Resource reclaim for sandboxes, tunnels and sessions past their TTL or plainly idle. Catalog reconciliation against what agents report is really running. Scorecard remediation for a component that dropped below its tier. Finding triage that proposes a severity and an owner drawn from catalog ownership. The cadence on top of that is also designed and not built: an opt-in, default-off sweep per workspace, with a credit budget it will not exceed, dormant workspaces excluded so it does not spend money watching nothing, duplicate proposals suppressed while one is still open, and a kill switch. The rule for the first version is unambiguous — there is no unattended-execute path. Everything the loop produces is a plan waiting for a person. We are describing this before it ships because the sequencing is the point. Plenty of tools add autonomy first and governance later. VibeControls built the review gate, the ledger, the identity boundary and the budget first; the decisions are what comes next, and they will arrive behind all four. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Turn what you just debugged into reusable AI context ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Shared AI prompt library with project context — VibeControls Use Cases > Create a shared AI prompt library with reusable project context, variables, categories, tags, and workspace sharing. Source: https://vibecontrols.com/use-cases/ai-prompt-library ## Shared AI prompt library with project context Save, tag, and reuse prompts with shared project context Teams waste hours reformulating the same AI prompts — "review this code for security issues", "generate test cases for this API endpoint", "explain this function" — from scratch every day, usually without the project context the AI needs to give useful answers. VibeControls AI Prompts solves this in two steps. First, create an AI Context: paste in your project's README, architecture diagram, coding standards, or any background the AI should know. Name it "My Project Context" and save. Then create a Prompt template: the actual instruction with {{variable}} placeholders for anything that changes run-to-run. Attach your Context, choose a supported category such as Review, Testing, Documentation, or Deployment, and add tags that help your team find it. Now every workspace member with access can find the prompt in the library and reuse its variables and shared context. To dispatch it, select an online agent with an AI-provider plugin configured for the work; the rendered prompt carries the selected values and attached context. No chat history juggling, no repeated "here's how our codebase works" preamble. Prompt categories and tags make it easy to find the right template. Dispatch is intentionally agent-backed today; an in-app queue for overnight batch runs is not yet available. ### Do it yourself Build a shared, project-aware AI prompt library: create reusable context, write tagged prompt templates with variables, share them workspace-wide, and dispatch them through an online agent with an AI provider. - 1 Create an AI Context /vibecontrols/ai-tools/ai/contexts Open the Contexts tab and add a reusable context from your project material, such as a README, architecture notes, or coding standards. You should see: Your project context is saved and ready to attach to any prompt, so every run carries the right background. [Open in app](https://app.vibecontrols.com/vibecontrols/ai-tools/ai/contexts) Mark done - 2 Create a prompt template /vibecontrols/ai-tools/ai/prompts Switch to the Prompts tab, add a new prompt, and write your template using {{variable}} placeholders wherever you want fill-in-the-blanks. - 3 Tag the prompt by category /vibecontrols/ai-tools/ai/prompts In the prompt editor, pick a supported category such as Review, Testing, Documentation, or Deployment, and add tags that help your team find it. - 4 Attach context and share workspace-wide /vibecontrols/ai-tools/ai/prompts Attach the AI Context you created to the prompt, then flip the Shared toggle on. - 5 Fill in variables and dispatch through an online agent /vibecontrols/ai-tools/ai/prompts Open the prompt, fill in the values for each {{variable}}, select an online agent with an AI-provider plugin, and click Dispatch. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Instant public URL for your local dev server Send code to AI directly from VS Code ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. --- # Turn what you just debugged into reusable AI context — VibeControls Use Cases > Capture and reuse AI contexts from terminal sessions. Share prompts across teams and integrate them into harness chats or workbench drawers. Source: https://vibecontrols.com/use-cases/ai-session-context-capture ## Turn what you just debugged into reusable AI context Author contexts and prompts against the live session, then insert them into any chat The most valuable context your team has is created during debugging and then thrown away. The command that finally reproduced the bug, the API call that returned the surprising payload, the paragraph of explanation you typed into a chat window — all of it lives in scrollback until the scrollback is gone. VibeControls makes that context a workspace object. An AI Context has a type — git repo, API call, markdown doc, command, plain text, file or URL — plus tags, a status of active or archived, and optional provenance: the session it was authored against and the Vibe it belongs to. Vibe-linked contexts carry a visibility of private or shared-with-vibe, so a note you make while debugging can stay yours or become the project's. Prompts work the same way but add structure. Each one is a template with {{variable}} placeholders, filed under a category — coding, debugging, review, documentation, testing, deployment, runbook or general — with a draft/active/archived lifecycle. Rendering happens on the server, so filling the variables produces exactly the same text for everyone. Usage is counted and the last-used time recorded, which makes it obvious which templates the team actually relies on and which were written once and forgotten. Using them is deliberately close to the work. In the AI workbench, the composer's insert picker lists your prompts and contexts, renders a prompt's variables inline, and drops the result into the message. On a session detail page, the workbench drawer opens beside the terminal with notes, prompts and history tabs, so a runbook prompt is one click from the shell you are already in. Files can be pulled off the agent straight into a context, and instruction files can be pushed down to the agent's working directory so the harness reads them on its next run. One honest characteristic: this is curated retrieval, not a vector index. There is no embedding store quietly deciding what the model sees — you choose, and what you chose is inspectable afterwards. That is cheaper and far easier to audit, and it is exactly why the separate signed per-Vibe context URL exists for the cases where an external tool needs a defined slice of the project instead. ### Do it yourself Capture the context and prompts a debugging session produced, scope them to a Vibe, and reuse them from the composer picker or beside the terminal. - 1 Capture the context you just built /vibecontrols/ai-tools/ai/contexts Open Contexts and create one: choose the type (git repo, API call, markdown doc, command, plain text, file or URL), paste the content, and tag it. You should see: The context is stored in the workspace with its type, tags and status. [Open in app](https://app.vibecontrols.com/vibecontrols/ai-tools/ai/contexts) Mark done - 2 Scope it to the project /vibecontrols/ai-tools/ai/contexts Link the context to a Vibe and set its visibility to private or shared with the Vibe. - 3 Write the prompt template /vibecontrols/ai-tools/ai/prompts Create a prompt with {{variable}} placeholders, pick a category such as debugging or runbook, and save it as active. - 4 Push the standing instructions to the agent /vibecontrols/ai-tools/ai/instructions Open Instructions and write or update the instruction file for the agent or the whole Vibe. - 5 Insert it into a harness chat /vibecontrols/ai Back in the AI workbench, use the composer's insert picker to pull in a prompt or context and fill in its variables. - 6 Reach the same library from the terminal /vibecontrols/sessions/:id Open a session and use the workbench drawer beside the terminal to browse notes, prompts and history, then copy the one you need. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling AI spend you can see — checked before the turn, charged after it AI that proposes the ops fix and waits for your approval ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # See exactly what your AI agents did — task by task — VibeControls Use Cases > Track AI agent tasks in real-time with VibeControls. View task status, source tool, and agent details in a live stream for full transparency. Source: https://vibecontrols.com/use-cases/ai-tool-activity-feed ## See exactly what your AI agents did — task by task A live activity feed of every AI tool call across Claude Code, Cursor, opencode and more — completions, failures, and payloads in one place When a team runs AI coding tools at scale, the hardest question to answer is a simple one: what did the agents actually do? Each tool keeps its own local history, on its own machine, in its own format. There is no shared place to see that Claude Code finished a task on one box while Cursor failed on another. VibeControls closes that gap with an AI tool-call activity feed. Every AI coding tool running on a registered agent emits events — task.started, task.completed, task.failed — tagged with the source tool (opencode, claude-code, gemini-cli, cursor, or any tool you wire up) and carrying the full JSON payload of what happened. Those events land in the workspace as first-class records you can query and watch. Because the feed is a live subscription, activity shows up as it happens: open the app and a completion on a remote agent surfaces as a notification the moment the tool reports it. A failed task announces itself instead of hiding in a log on a machine nobody is looking at. You can filter by agent, by source tool, or by time window to answer "what ran overnight?" or "which tool keeps failing on the staging box?" This is observability for AI work, and it is deliberately separate from the human audit log. The audit log answers "which person changed what"; the tool-call feed answers "which AI tool did what, on which agent, and did it succeed." Every event carries its operation context, so it ties back to the right workspace and project. For an engineering manager standing up an AI-native team, this is the difference between trusting the fleet and guessing at it. You get a task-by-task record of agent work across every tool, retained on your terms — old events can be pruned on a schedule — and visible from one workspace instead of scattered across a dozen terminals. ### Do it yourself Watch a live, task-by-task feed of what your AI coding tools did across the fleet — completions and failures by source tool and agent — and filter the history when you need it. - 1 Run your AI coding tools on registered agents /vibecontrols/agents Use Claude Code, Cursor, opencode, Gemini CLI, or your own tool on a machine registered as a VibeControls agent. You should see: Each tool emits task.started / task.completed / task.failed events tagged by source and agent. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 See activity arrive live /vibecontrols Keep the app open — the feed is a live subscription, so nothing to click. - 3 Open an agent to review its recent tool activity /vibecontrols/agents Open the agent you want to inspect to see the AI tool events streaming from it. - 4 Filter the AI tool events in the audit view /vibecontrols/audit Open the audit log and filter the resource type to AI Tool Event to see the recorded history. - 5 Prune old events on your terms /vibecontrols Set a retention cutoff so tool events older than a chosen date are cleared. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling One source of truth for your AGENTS.md and .cursorrules Scan every vibe and agent — then triage the findings in one queue ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Full audit trail for every operation, every actor, every outcome — VibeControls Use Cases > Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Source: https://vibecontrols.com/use-cases/audit-log-accountability ## Full audit trail for every operation, every actor, every outcome "Who restarted the database at 3 AM?" answered in 10 seconds A post-incident review without a complete audit trail is guesswork. "Someone restarted the service" is not a useful finding. "Aditya Kumar ran the Restart API tile on the staging VibeDeck at 03:12 UTC, from IP 203.0.113.45, exit code 0" is. VibeControls captures operations across the platform in a tamper-evident audit log. Session starts and stops, VibeDeck tile executions, configuration changes, webhook dispatches, scheduled job runs — logged with actor type (user, app, service, super_admin), actor identity, target resource, result, and timestamp. The audit log is retained for 7 years with immutable batch logging; no plan tier or workspace admin can shorten the window. It is searchable and filterable from the Workspace Settings → Audit view. Filter by actor to see everything one engineer did in a timeframe. Filter by resource type to see all mutations on a particular agent. Export as JSON or CSV for SIEM ingestion or compliance reporting. The per-operation detail shows full execution output, not just exit code. When the 3 AM restart audit trail shows success but the service is still down, you have the output to tell you why. Investigations that used to take 45 minutes of log-digging take 10 seconds. ### Do it yourself Investigate any operation in your workspace by opening the audit log, filtering it down to the exact actor or resource, inspecting full execution detail, and exporting the results for the record. - 1 Open the Audit page /vibecontrols/audit From your workspace, head to the Audit page to see the full trail of activity. You should see: The audit log table loads alongside summary stats showing every captured operation across your workspace. [Open in app](https://app.vibecontrols.com/vibecontrols/audit) Mark done - 2 Search the trail /vibecontrols/audit Type any keyword — a command, a resource name, an actor — into the search box at the top of the page. - 3 Filter by action, resource, or status /vibecontrols/audit Use the filter bar to pick an action type, a resource type (such as Session or Deck Button), or an outcome like Success or Failure. - 4 Filter by actor /vibecontrols/audit Choose a specific actor — an engineer, agent, vibe, or session — from the actor filter to see everything that one identity did. - 5 Open an entry's full detail /vibecontrols/audit Click any row in the table to expand its full detail. - 6 Verify operations were captured /vibecontrols/audit Filter the resource type to confirm specific events were logged — for example set it to Deck Button to find a VibeDeck tile run, or Session to find a session start. - 7 Export the results /vibecontrols/audit Click the Export action and choose CSV or JSON to download exactly the filtered set you're viewing. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs Quality scorecards your CI keeps honest Enterprise SSO via OIDC — every engineer logs in with their corporate identity ### Related use cases Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. Enterprise & Governance ### Connect AI coding tools to an agent-local plan workspace Set up the agent-local Plannotator integration for supported AI CLIs, then use its embedded workspace to inspect the plan sessions those tools create. --- # Give external beta testers time-limited API access via tunnel token — VibeControls Use Cases > Provide beta testers with time-limited API access. Create a public-access tunnel with a 4-hour expiry for testing. Source: https://vibecontrols.com/use-cases/beta-tester-tunnel ## Give external beta testers time-limited API access via tunnel token Create a 4-hour public tunnel URL — no VPN, no SSH, auto-expires Getting external beta testers to access a staging API is usually a pain. You either make the server public (security liability), give them VPN credentials (setup overhead), or ask them to coordinate around a specific time window when you'll open access manually. VibeControls tunnel tokens provide a clean third path. The staging API runs on a server with an agent installed. Create a port-forward to the API port and enable "Public access". Set the expiry: 4 hours for a focused test session, 24 hours for a day-long integration sprint. Copy the public URL and send it to the beta testers. They can hit the API from any browser, Postman, or curl command using the URL. No credentials needed — the tunnel token is the access control. When the expiry time arrives, the URL stops working automatically. No cleanup, no "remember to revoke the access we granted last week." The access lifecycle is completely managed by the token. For QA teams running external user acceptance testing, this reduces coordination overhead from hours (set up VPN, share credentials, verify connectivity, clean up) to minutes (create token, share URL, forget about it). ### Do it yourself Give external beta testers a public URL to your staging API that automatically stops working after a time window you choose, with no manual cleanup. - 1 Open your agent /vibecontrols/agents Go to Agents and open the agent running on the box that hosts your staging API. You should see: The agent detail page opens, ready to forward one of its ports. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 Create a port forward to your API /vibecontrols/tunnels/new Click New tunnel, pick HTTP, and point it at the port your staging API listens on. - 3 Start the tunnel /vibecontrols/tunnels Open the tunnel you just created and click Start so it begins serving. - 4 Grant time-limited public access /vibecontrols/tunnels On the tunnel's Overview tab, open the Time-limited public access panel and grant a window — 1 hour, 4 hours, 24 hours, or 7 days. - 5 Copy the public URL and send it to testers /vibecontrols/tunnels Copy the public URL shown in the same panel and share it with your beta testers — no credentials to hand out. - 6 Let it expire — or revoke early /vibecontrols/tunnels Do nothing and the URL stops serving automatically when the deadline passes; or click Revoke now to cut access immediately. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Review AI-generated scripts in an isolated sandbox before real execution A workspace AI assistant that acts as you — never beyond you ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # Any machine's terminal in a browser tab — VibeControls Use Cases > Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. Source: https://vibecontrols.com/use-cases/browser-terminal ## Any machine's terminal in a browser tab Full interactive terminal access without installing an SSH client You're on your work laptop, and you need to check a log on the staging server — but your SSH config lives on your home machine. Or you're at a conference on a borrowed laptop with no tooling installed. Or you just want to avoid switching windows. VibeControls browser terminals solve all three scenarios. Every session you start on any registered agent becomes accessible via a browser URL. Click "Open Terminal" and a full, interactive ttyd-powered terminal opens in a new tab — fully functional, copy-paste intact, keyboard shortcuts working. Sessions are persistent. Close the tab and come back later — the session is still running, still in the same state. tmux, Zellij, and WezTerm provider plugins mean your layout and split-panes survive reconnects too. For teams, this removes the "SSH into the box" prerequisite from every runbook. Anyone with the right workspace role can access the terminal without needing to know the IP, port, or SSH key — VibeControls handles the authentication at the workspace edge. ### Do it yourself Open a fully interactive, persistent terminal for any of your registered machines right inside a browser tab — no SSH required. - 1 Confirm your agent is online /vibecontrols/agents/:id Open the Agents page and pick the machine you want to reach. Make sure it shows as healthy and reachable so it can host your browser terminal. You should see: The agent appears as ready, with its default terminal provider available. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 Create a session /vibecontrols/sessions Go to Sessions and click New. Pick your agent and start the session. - 3 Open the terminal in a tab /vibecontrols/sessions On the session card, click Open Terminal. - 4 Work as if you were SSH'd in /vibecontrols/sessions Type commands, copy and paste, and use your keyboard shortcuts. Everything behaves like a real terminal. - 5 Close the tab and come back anytime /vibecontrols/sessions Close the terminal tab whenever you like, then click Open Terminal again later to reattach. - 6 Keep your tmux, Zellij, or WezTerm layout /vibecontrols/sessions Use your favorite multiplexer inside the session as you normally would. - 7 Let teammates open it without SSH /vibecontrols/sessions Share access through your workspace roles. Teammates with the right role can open the same terminal from their own browser. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Per-project environment variables without .env sprawl Instant public URL for your local dev server ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Instant public URL for your local dev server Share your local dev server via a public HTTPS URL instantly. No VPN or staging needed—just two clicks to share with your team or clients. --- # Bring your own model keys — and keep them on your own machine — VibeControls Use Cases > Securely store API keys on your machine. VibeControls supports various providers without needing to access your keys for AI harness sessions. Source: https://vibecontrols.com/use-cases/byok-ai-provider-keys ## Bring your own model keys — and keep them on your own machine Provider credentials live in the agent's config, not in our platform Most AI developer tools ask you to hand over a provider key, or quietly resell inference through their own account. Either way your prompts and your spend route through someone else's infrastructure, and you find out what that means only when something goes wrong. VibeControls draws the line somewhere else. Open the AI Keys panel on an agent and you get a row per provider whose plugin reads credentials from that agent's own config: Anthropic (API key or auth token), OpenAI Codex, Google Gemini (either GEMINI_API_KEY or an existing GOOGLE_API_KEY), Cursor, OpenRouter, Minimax, Ollama, and a generic OpenAI-compatible slot that takes both a key and a base URL so Together, Groq or Mistral work without a new plugin. Secret fields are write-only once saved. Two of those entries matter more than they look. The OpenAI-compatible slot means any endpoint that speaks the OpenAI protocol is a first-class provider, including one you host yourself. The Ollama entry takes an optional host override, so a local model server or a private GPU box is a normal choice in the model picker rather than a workaround. Harnesses that authenticate through their own binary — Aider, Amp, Copilot, Crush, Goose, Plandex, Pi — never appear in the keys panel at all, because their credentials live in their own config directories on disk. Bridle manages those files instead. Nothing about that path involves the platform either. The honest boundary: the built-in workspace AI Assistant is a different surface. It runs in a per-conversation sandbox on workspace compute and is metered in AI credits, because you are not supplying the key there. Harness sessions spend your provider key and are not metered in credits. That split is deliberate — it lets you decide, per surface, whether the inference is yours or ours, and it means VibeControls is not in the business of reselling tokens. ### Do it yourself Put your own provider keys on the agent — including an OpenAI-compatible base URL or a self-hosted Ollama host — so harness sessions run on your inference, not ours. - 1 Install the provider plugin /vibecontrols/agents/:id On the agent, open the Plugins tab and install the provider plugin for the model vendor you want to use. You should see: The provider appears as an available harness for that agent. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 Open the AI Keys panel /vibecontrols/ai In the AI workbench, open the AI Keys panel from the admin bar. Each provider has its own row with the exact config keys its SDK reads. - 3 Save the credential on the agent /vibecontrols/ai Paste the API key into the provider row and save. For an OpenAI-compatible provider, add the base URL too; for Ollama, optionally set the host. - 4 Point at a self-hosted or aggregated model /vibecontrols/ai Start a session and open the model selector; the models listed are the ones your configured provider reports. - 5 Manage the CLI-only harnesses separately /vibecontrols/ai/config For harnesses that authenticate through their own binary, open the AI configuration view and use Bridle to manage their config files on disk. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Run Claude Code, Codex, Gemini and OpenCode from one workbench Decide how much rope the agent gets — per session, per machine ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. --- # The entire platform from your terminal — and your CI pipeline — VibeControls Use Cases > Access the full VibeControls platform via CLI. Automate tasks with scriptable output and CI-ready authentication. Source: https://vibecontrols.com/use-cases/cli-automation ## The entire platform from your terminal — and your CI pipeline One binary, 23 command modules, device-code login or API keys for automation Web dashboards are great until you need to automate. The moment a workflow belongs in a deploy script or a cron job, teams start scripting against raw APIs with hand-rolled auth — and every script becomes a small, unaudited bypass of the platform. The VibeControls CLI closes that gap. Run "vibecontrols auth login" and it prints a device-code URL — approve it in any browser, on any machine, and the token is stored locally under ~/.config/vibecontrols-cli/. Run "vibecontrols workspace set " once and every subsequent command is scoped to that workspace. From there, the full surface is mounted as 23 command modules: the vibes/agents/sessions core, sandboxes, RBAC, tags, files, scheduler, notifications, feature flags, billing, devportal, health, and more. "vibecontrols vibes list" returns the same vibe set you'd see in the web app, in machine-readable output. "vibecontrols rbac check read project" prints a permission decision you can use as a gate in a pipeline. For CI, skip the browser entirely: mint an API key in the devportal and the same commands run headless. "vibecontrols sandbox create" in a deploy script provisions an isolated environment and exits non-zero on failure, so your pipeline reacts correctly. Revoke the key in the devportal and CI access is cut immediately. Crucially, there is no CLI bypass. Every command goes through the same public gateways, the same RBAC enforcement, and the same quotas as the web UI — and lands in the same audit log. Automation uses the governed surface, not a side door. ### Do it yourself Drive the entire VibeControls platform from your terminal and CI pipeline — logging in, scoping a workspace, listing vibes, provisioning sandboxes headlessly, and gating deploys on RBAC checks, all through the same gateways, quotas, and audit as the web app. - 1 Install the CLI Terminal Install the VibeControls CLI globally so the `vibecontrols` command is available everywhere on your machine. @@MD@@``` npm i -g @vibecontrols/cli ```@@MD@@ You should see: One binary with all 23 command modules is installed and ready to run. Mark done - 2 Log in from your terminal Terminal Run the login command. It prints a device-code URL and a short user code, then opens your browser so you can confirm. - 3 Scope your workspace Terminal Point the CLI at the workspace you want to work in by passing its ID. - 4 List your vibes Terminal Pull the same set of vibes you see in the web app, now in clean machine-readable output you can pipe into scripts. - 5 Mint an API key for CI /devportal Open the developer portal and generate a headless API key for your pipeline. Copy the secret — it is shown only once. - 6 Provision a sandbox headlessly in CI Terminal Using the API key from the devportal, create a sandbox environment directly inside your pipeline. - 7 Gate the deploy on an RBAC check Terminal Ask the CLI whether a given user is allowed to perform an action, and use the result as a pass/fail gate in your pipeline. - 8 Revoke the key when you are done /devportal Back in the developer portal, revoke the CI API key once the pipeline run is complete or the key is no longer needed. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Roll out new capabilities to your whole fleet from the plugin catalog New hire productive in under an hour ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Living software catalog with real ownership and dependency graphs — VibeControls Use Cases > Maintain a live software catalog with VibeControls. Register services, APIs, and libraries with ownership and dependency graphs. Source: https://vibecontrols.com/use-cases/component-catalog ## Living software catalog with real ownership and dependency graphs Every service, API, and library registered — with owners, docs, and live health Every growing engineering org eventually builds a spreadsheet of services. Then someone updates the spreadsheet. Then someone else doesn't. Six months later, the spreadsheet is the least reliable source of truth in the organization. VibeControls Catalog is different because it's connected to the work surface. Services are registered as typed Component entities (Service, API, Library, Resource, Job, Workflow) with ownership, lifecycle state, documentation links, and dependency relationships. Components can be linked to the active Vibes where developers are working on them. When an engineer updates a service's Vibe — changes environment variables, updates documentation, runs actions — the catalog entry reflects the living state of the service. The dependency graph shows at a glance how services relate: which APIs call which backends, which libraries are shared. Impact analysis becomes possible: "if I change this API contract, which downstream services does it affect?" The catalog answers that question from the dependency graph — not from asking everyone in a Slack thread. Lifecycle states (active, deprecated, retired) give your org a structured way to manage technical debt. ### Do it yourself Register your services as typed, owned components with lifecycle states, map their dependencies on a live graph, attach docs and quality scorecards, and organize everything into a shared catalog. - 1 Register a component /vibecontrols/components Open Components and click Create. Give your service a name and pick its entity type — Service, API, Library, Resource, Job, or Workflow. You should see: A new typed component appears in your catalog with the kind you selected. [Open in app](https://app.vibecontrols.com/vibecontrols/components) Mark done - 2 Set the owner and lifecycle /vibecontrols/components In the component form, assign the owning team or person and choose a lifecycle stage — active, deprecated, retired, or experimental. - 3 Add documentation links /vibecontrols/components Open the Docs tab and add the repository URL plus a docs site for the component. - 4 Define dependencies /vibecontrols/components Open the component's Dependencies tab and add the other components it relies on, choosing the relationship type for each edge. - 5 View the dependency graph /vibecontrols/components Stay on the Dependencies tab to see the interactive graph, with your component as the root node and its dependencies branching off, alongside a table. - 6 Check the quality scorecard /vibecontrols/components Open the Scorecard tab on the component to review its quality scores against your standards. - 7 Organize into a catalog /vibecontrols/catalogs/:id Open a catalog and add your components and related Vibes to it. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Platform team's shared ops deck eliminates 80% of tickets Auto-built docs site embedded in your project Vibe ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # Connect your database GUI to a remote database over a tunnel — VibeControls Use Cases > Connect your database GUI to a remote database using VibeTunnels. No VPN or IP allow-listing needed for seamless access. Source: https://vibecontrols.com/use-cases/database-port-forward ## Connect your database GUI to a remote database over a tunnel TablePlus, DataGrip, or psql — against a remote DB, no VPN required Database GUIs are designed for localhost. Getting them to work against a remote database usually means either opening the database port to the world (insecure), configuring VPN (complex), or doing everything over an SSH tunnel with a manually-maintained config. VibeControls makes raw-TCP tunneling a first-class operation. Register the database server as a target, install an agent on it, and create a TCP tunnel: the agent forwards its local database port out to a public VibeTunnels endpoint. VibeControls routes the tunnel through VibeTunnels' frpc infrastructure and gives you a public endpoint — an assigned *.vibetunnels.com host and port. Open TablePlus or DataGrip, point it at that host:port, and connect. The GUI talks to the remote database transparently. The tunnel is a workspace resource: it's listed in your dashboard, has a creation timestamp, and can be paused or deleted when you're done. No more SSH config entries to maintain, no more "what was that port again?" — the tunnel is there when you need it and gone when you don't. ### Do it yourself Tunnel a remote database through VibeControls so you can connect your database GUI to it over a secure, time-limited public endpoint. - 1 Confirm your agent is running on the database server /vibecontrols/agents Open Agents and check that the agent installed on your database server is online. Open its detail page to confirm the VibeTunnels provider is available. You should see: Your database-server agent shows as connected, with the VibeTunnels provider ready to host TCP tunnels. [Open in app](https://app.vibecontrols.com/vibecontrols/agents) Mark done - 2 Start a new tunnel /vibecontrols/tunnels/new Go to Tunnels and click New to open the tunnel creation wizard. - 3 Pick the VibeTunnels provider and your DB agent /vibecontrols/tunnels/new In step 1, select the VibeTunnels provider and choose the agent running on your database server. - 4 Choose TCP and point it at your database /vibecontrols/tunnels/new Select the TCP protocol and enter your database port, 5432. - 5 Create and start the tunnel /vibecontrols/tunnels/new Submit the wizard to create and start the tunnel. - 6 Grant a time-limited public window /vibecontrols/tunnels Open the tunnel's Overview tab and grant public access for a preset window (1h, 4h, 24h, or 7d), then copy the public URL. - 7 Connect your database GUI Terminal Open TablePlus or DataGrip and create a new database connection using the *.vibetunnels.com host and assigned port from the tunnel, with your usual database credentials. - 8 Pause or delete the tunnel when done /vibecontrols/tunnels From the tunnel's detail page, pause the tunnel to stop access, or delete it to remove it entirely. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Self-serve operations for the whole org via shared VibeDeck Instant Slack alert when a session fails ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Auto-built docs site embedded in your project Vibe — VibeControls Use Cases > Embed an auto-built docs site in your project Vibe. Use Markdown for documentation and eliminate the need for separate platforms. Source: https://vibecontrols.com/use-cases/docs-site-in-vibe ## Auto-built docs site embedded in your project Vibe Markdown-powered docs, compiled asynchronously, visible right where work happens Documentation lives in three places: a Confluence nobody updates, a GitHub wiki that's two major versions behind, and the heads of the senior engineers who wrote the system. None of these is where developers actually work. VibeControls Docs brings documentation to the Vibe — the same place where sessions, environment variables, and actions live. Create a docs site for a Vibe, then author its pages in the built-in Markdown editor. VibeControls compiles the site asynchronously in the background. The resulting documentation site is embedded in the Vibe detail view. Anyone who opens the Vibe sees the current, compiled docs immediately — no separate URL to remember, no confluence login, no wiki drift. The async builder means you don't block on a compilation step. Make a change to a Markdown page, trigger a rebuild, and the embedded docs update without leaving the Vibe dashboard. Each Vibe currently has one managed docs site, with page trees, build revisions, and rollback for keeping user-facing docs, API reference, and internal runbooks together. ### Do it yourself Build one managed Markdown documentation site for a Vibe, compile it asynchronously, and embed it directly alongside the project work. - 1 Open the Vibe's Docs tab /vibecontrols/vibes Open your project Vibe and click the "Docs" tab at the top of the detail view. You should see: The Docs panel for your Vibe opens, ready for you to set up a documentation site. [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Create a docs site /vibecontrols/vibes Click "Create docs site" and pick a supported static-site generator (Docusaurus, MkDocs, or VitePress). - 3 Author your Markdown pages /vibecontrols/vibes Use the page tree to add pages, then write each one in the Markdown editor and check the live preview beside it. - 4 Compile the site /vibecontrols/vibes Click "Compile" (turn on "publish" if you want it live) to kick off the build in the background. - 5 View the embedded site /vibecontrols/vibes Once the build succeeds, browse the compiled site right in the Docs panel, or open the full-screen viewer. - 6 Rebuild after edits /vibecontrols/vibes Edit any Markdown page and click "Compile" again to publish a fresh version; use the revision history to roll back if needed. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Living software catalog with real ownership and dependency graphs See exactly what version is deployed in every environment ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # See exactly what version is deployed in every environment — VibeControls Use Cases > Monitor deployed versions across environments with the Environment Topology view. Ensure safe promotions with a full audit trail. Source: https://vibecontrols.com/use-cases/env-topology-view ## See exactly what version is deployed in every environment Dev, staging, prod — one view, live deployment state across all services "What version is in staging?" is a question that should have a one-second answer. In practice, it usually involves checking a CI/CD dashboard, correlating with a Git tag, and hoping the label matches the binary. VibeControls' Environment Topology view makes version state a first-class platform concern. Every service is tracked across its environment lifecycle — dev, staging, and production. The current version in each environment is recorded whenever a deployment is registered with the platform. The topology view is a grid: services on one axis, environments on the other. Yellow means staging is behind production. Red means prod hasn't received a release that's been in staging for a week. Green across the board means everything is in sync. When you're ready to promote, the platform records the promotion event in the audit log: who triggered it, which version moved, from which environment to which. The approval gate (for platform teams that require sign-off before production) is built in. This turns "what's deployed where" from a question answered by tribal knowledge into a dashboard answered in one glance. ### Do it yourself See the live version of every service across development, staging, and production in one matrix, spot version drift at a glance, and safely promote a version forward with a built-in audit trail. - 1 Open the Environment Topology view /vibecontrols/environments From the VibeControls sidebar, click the Environment Topology nav item to open the environments page. You should see: A grid loads with your services listed down the side and Development, Staging, and Production across the top. [Open in app](https://app.vibecontrols.com/vibecontrols/environments) Mark done - 2 Read the deployed version in every cell /vibecontrols/environments Scan across each service row to see the version currently running in Development, Staging, and Production. - 3 Spot version drift at a glance /vibecontrols/environments Look at the cell colors: green cells match what is in production, amber cells differ from production. - 4 Record a deployment /vibecontrols/environments Click "Record deployment", pick the service and environment, and enter the version that is now live there. - 5 Promote a version forward /vibecontrols/environments Click "Promote version", choose the service, the source environment, and the target environment to move that version into. - 6 Confirm the promotion and recheck for drift /vibecontrols/environments Review the grid after promoting to confirm the target environment now shows the expected version. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Auto-built docs site embedded in your project Vibe Pair on a live terminal session from a Slack link ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # Golden-path templates that preserve service conventions — VibeControls Use Cases > Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Source: https://vibecontrols.com/use-cases/golden-path-template ## Golden-path templates that preserve service conventions Publish component snapshots, variables, and setup guidance for consistent starts Every new microservice at a growing org reimplements the same 80%: the ownership model, service metadata, setup guidance, and a starting repository. If those conventions are not captured somewhere, every team does them slightly differently — and months later you are dealing with a patchwork of approaches. VibeControls Templates lets you capture a component snapshot with a category, version, descriptive file tree, variables, setup instructions, and an optional source repository. Publish that template to the workspace catalog so teams can start from the same metadata and conventions. Now a new team picks it from the gallery, fills in the requested parameters, and instantiates a component from the frozen snapshot. When a source repository is configured, the flow can create or fork a GitHub repository and add the template's rendered README and setup guidance. The catalog retains the template relationship for instantiated components. Automated generation of Dockerfiles, CI, VibeDeck buttons, security configuration, and cross-version drift reporting are planned enhancements, not part of the current template flow. ### Do it yourself Package a component snapshot as a reusable template, publish it to the catalog, and let teams instantiate consistent components with requested parameters and setup guidance. - 1 Publish your component as a template /vibecontrols/templates Open Templates, choose Publish, and pick an existing component as the source. Give it a name, category, and version. You should see: VibeControls freezes a snapshot of your component and creates a draft template tied to that exact version. [Open in app](https://app.vibecontrols.com/vibecontrols/templates) Mark done - 2 Capture setup guidance and conventions /vibecontrols/templates In the publish form, record the descriptive file tree, setup instructions, optional source repository, and the conventions a team should follow. - 3 Define the parameters teams will fill in /vibecontrols/templates Add the variables a new team will provide (for example service name, port, and owner), then publish the template to the catalog. - 4 Browse the gallery and pick a template /vibecontrols/templates A new team opens Templates and selects the 'Node API Microservice' template from the gallery. - 5 Fill in the three parameters /vibecontrols/templates In the wizard, enter the requested values for each parameter and submit. - 6 Optionally create or fork a repository /vibecontrols/templates When the template includes a source repository, choose the GitHub repository option from the instantiation flow. - 7 Review template provenance /vibecontrols/templates Return to the template's detail page and review the components instantiated from it. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams New hire productive in under an hour Platform team's shared ops deck eliminates 80% of tickets ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. Team & Platform ### Living software catalog with real ownership and dependency graphs Maintain a live software catalog with VibeControls. Register services, APIs, and libraries with ownership and dependency graphs.'s connected to the Vibes where work actually happens. --- # Pair on a live terminal session from a Slack link — VibeControls Use Cases > Share terminal sessions via Slack with teammates. Collaborators can join read-only or request control. All actions are audited. Source: https://vibecontrols.com/use-cases/live-session-share ## Pair on a live terminal session from a Slack link Expiring share links, viewer or interactive roles, and keyboard handoff in one click Remote pairing over screen share is a study in lag: the driver types, the watcher squints at compressed pixels, and "wait, scroll up" costs thirty seconds each time. And when a teammate just needs to see your terminal for two minutes, provisioning real access is overkill. VibeControls session sharing fixes both. From any running session, open the Share dialog and pick a path: grant access to named workspace members with a viewer or interactive role, or create a share link with an expiry, an optional password, and usage limits. Drop the link in Slack. Your teammate clicks it, lands on the shared-session page, and joins as a viewer — you see them arrive in the collaborator list in real time. Both of you are looking at the same live terminal buffer, not a video of it. No screen-share lag, no "can you zoom in", and viewers physically cannot type. When the watcher needs to drive, they click "Request control". You get prompted, accept, and the keyboard is theirs — control changes propagate to both sides in about a second. Take it back whenever you want, or release it explicitly. If a collaborator disconnects mid-control, control reverts to you automatically. Done debugging? Revoke the link and the token stops validating immediately — no lingering access. Recipients see everything shared with them in one "shared with me" list, the same flows work from the VS Code extension, and every share, join, and control transfer lands in the workspace audit log. Pairing, incident response, code review walkthroughs, onboarding demos — one mechanism covers them all. ### Do it yourself Share a running terminal session as a password-protectable, expiring link so a teammate can join live, watch in real time, and request keyboard control with a handoff in about a second. - 1 Open your running session /vibecontrols/sessions/:id Go to the session you want to pair on and open its detail page so you can see the live terminal. You should see: Your live terminal session is open and ready to share. [Open in app](https://app.vibecontrols.com/vibecontrols/sessions) Mark done - 2 Open the Share panel /vibecontrols/sessions/:id Click the Share button on the session detail page to open the share-link panel. - 3 Invite a teammate by name with a role /vibecontrols/sessions/:id In the Specific people tab, pick a workspace member and choose whether they join as a viewer or with interactive access. - 4 Create an expiring, password-protected link /vibecontrols/sessions/:id Switch to the public link tab, set the viewer or interactive permission, add an expiry, an optional password, and an optional max-uses cap, then create the link. - 5 Teammate clicks the Slack link and joins /vibecontrols/share/:token Your teammate opens the link you shared and, if you set one, enters the password. - 6 Hand off keyboard control /vibecontrols/share/:token When your teammate clicks Request control, accept the request from the control bar to pass them the keyboard. - 7 Revoke the link when you're done /vibecontrols/sessions/:id Back in the share panel, find the active link and click Revoke. - 8 Find sessions shared with you /vibecontrols/sessions Open the Shared with me list to see every session a teammate has invited you to. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams See exactly what version is deployed in every environment Quality scorecards your CI keeps honest ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # Give your AI assistant scoped access to your codebase — VibeControls Use Cases > Generate scoped URLs for AI access to specific code files. Keep your full codebase private while providing necessary context to AI. Source: https://vibecontrols.com/use-cases/llm-context-url ## Give your AI assistant scoped access to your codebase Signed, expiring LLM context URLs — no raw API keys exposed AI coding assistants are most useful when they understand your entire project — but pointing them at a raw API key or a broad S3 bucket is a security anti-pattern. You want the AI to have context, not credentials. VibeControls' per-vibe LLM Context lets you configure exactly which files and metadata to expose: the README, the relevant service's source files, the architecture diagram, maybe the API schema. Nothing else. Click "Mint Token" and VibeControls generates a signed URL with a configurable expiry — one hour for a quick review session, 24 hours for an extended coding sprint. The URL is audited: every fetch shows up in your workspace's audit log with timestamp and IP. Paste the URL into Cursor's context panel, add it to your Claude Code CLAUDE.md, or pass it to any MCP-compatible AI agent. The AI reads your curated context. When the token expires, it stops working — no revocation ceremony needed. ### Do it yourself Generate a scoped, time-limited signed URL that exposes exactly the files and metadata you choose from a Vibe, paste it into your AI assistant, and watch every fetch get audited until the token expires. - 1 Open the Vibe's LLM Context tab /vibecontrols/vibes/:id Open the Vibe you want to share, then select the LLM Context tab. You should see: You see the LLM Context panel with the settings form, your tokens, and the fetch audit view. [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Choose what to include and set expiry /vibecontrols/vibes/:id In the settings form, pick exactly what to expose (README, source, architecture, API schema) and set how long access should last. - 3 Preview the assembled context /vibecontrols/vibes/:id Click Preview to see the exact context that will be served, toggling between JSON and Markdown. - 4 Mint a signed URL /vibecontrols/vibes/:id Click Mint, give the token a label, and confirm the expiry. - 5 Copy the URL into your AI assistant Claude Code Hit Copy and paste the URL into Cursor, your CLAUDE.md, or any MCP agent. - 6 Watch fetches get audited /vibecontrols/vibes/:id Return to the LLM Context tab and check the audit view for this token. - 7 Let it expire or revoke early /vibecontrols/vibes/:id Do nothing and the URL stops working at expiry, or use the panel actions to revoke or rotate it immediately. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Send code to AI directly from VS Code Bootstrap a multi-server agent fleet in under 5 minutes ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. --- # Zero-config local agent in under a minute — VibeControls Use Cases > Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Source: https://vibecontrols.com/use-cases/local-agent-setup ## Zero-config local agent in under a minute From nothing to a unified control plane with one command Every developer's machine is a tangle of tmux windows, shell history, and half-remembered SSH aliases. VibeControls collapses all of it into one place. Install the agent with a single command — the in-product installer is "curl -fsSL [https://app.vibecontrols.com/install](https://app.vibecontrols.com/install) | bash" (or install the package directly with "npm install -g @vibecontrols/agent" / "bun add -g @vibecontrols/agent"). Run "vibe start" and it prints a tunnel URL and an API key. Paste those into "Add Agent" in the VibeControls dashboard and moments later your laptop is a live, managed agent — with a health heartbeat, a plugin inventory, and a terminal you can open from any browser tab. Optionally install the VibeControls CLI separately ("npm install -g @vibecontrols/cli") for workspace management — "vibecontrols auth login" to authenticate, and the full 175+ subcommand suite for managing agents, vibes, sessions, and more across your team. The agent and CLI are separate packages with different roles: the agent runs locally on each machine, while the CLI is your control plane client. From the moment the agent is running you can start named sessions, expose local ports as public URLs, attach your current project as a Vibe, and route any AI provider through the workspace. No YAML, no .env juggling, no separate tooling for each concern. When you close the laptop lid, sessions persist. When you reopen, the agent re-registers automatically. Run "vibe autostart install" once and it survives reboots too. ### Do it yourself Turn your laptop into a live, managed VibeControls agent with a health heartbeat, plugin inventory, browser terminal, and public tunnels — in under a minute. - 1 Install the agent Terminal Open your terminal and install the VibeControls agent globally with a single command. @@MD@@``` npm install -g @vibecontrols/agent ```@@MD@@ You should see: The install finishes and the vibe command is available on your PATH. Mark done - 2 Start the agent Terminal Run the start command to boot the agent on your laptop. - 3 Add your agent in the app /vibecontrols/agents Go to the Agents page, click Add Agent, choose Add from tunnel, and paste in the tunnel URL and API key your terminal printed. - 4 Check the health heartbeat /vibecontrols/agents Open your new agent to view its detail page. - 5 Browse the plugin inventory /vibecontrols/agents On the agent detail page, open the Plugins tab and switch between Installed and Available. - 6 Open a browser terminal /vibecontrols/sessions Head to the Sessions page and click Open Terminal. - 7 Expose a local port /vibecontrols/tunnels/new Open the New Tunnel page and point it at a port running on your machine. - 8 Survive reboots Terminal Run the autostart command so the agent comes back on its own after a restart. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Per-project environment variables without .env sprawl ### Related use cases Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. Individual Developer ### Instant public URL for your local dev server Share your local dev server via a public HTTPS URL instantly. No VPN or staging needed—just two clicks to share with your team or clients. --- # Instant public URL for your local dev server — VibeControls Use Cases > Share your local dev server via a public HTTPS URL instantly. No VPN or staging needed—just two clicks to share with your team or clients. Source: https://vibecontrols.com/use-cases/local-port-share ## Instant public URL for your local dev server Share localhost:3000 with a designer or client — no VPN, no deployment Frontend reviews shouldn't require a staging pipeline. When you need a designer to see your in-progress UI or a client to approve a workflow, spinning up a staging deploy is overkill — and often breaks because production config doesn't match dev. VibeControls tunnels solve this with a two-click flow: open the Tunnels panel, click "New Port Forward", pick your local port. VibeControls hands you a *.trycloudflare.com public URL over HTTPS. Share the link. The designer opens it in any browser on any device and sees your actual live dev server — hot-reload, console errors, and all. No build step, no environment mismatch, no "why does it look different in staging?" You can generate a public-access token with a 4-hour expiry for short review sessions, or keep the tunnel open for longer collaboration. When you're done, delete the tunnel from the dashboard and the URL goes dead. Clean, simple, reversible. ### Do it yourself Turn any local dev server into a secure, shareable HTTPS link with a self-expiring public access window, then kill the link the moment you're done. - 1 Open the Tunnels panel /vibecontrols/tunnels Head to the Tunnels panel from the VibeControls dashboard. This is where all your port forwards live. You should see: You see your list of active tunnels and a button to create a new one. [Open in app](https://app.vibecontrols.com/vibecontrols/tunnels) Mark done - 2 Start a new port forward /vibecontrols/tunnels/new Click New Port Forward to launch the setup wizard, then pick Cloudflare as the provider and choose the agent running on the machine that hosts your dev server. - 3 Pick your local port /vibecontrols/tunnels/new Enter the local port your dev server is running on (for example 3000 or 5173). Optionally add a basic-auth username and password to lock down the URL. - 4 Get your public HTTPS URL /vibecontrols/tunnels/new Review the summary and submit to start the tunnel. - 5 Grant a time-limited public window /vibecontrols/tunnels Open the new tunnel and, in the Overview tab's public-access panel, grant public access with a 4-hour window (or pick 1 hour, 24 hours, or 7 days). Copy the URL and share it with your reviewers. - 6 End access whenever you want /vibecontrols/tunnels When the review is over, click Revoke in the same panel, or just delete the tunnel to tear it down entirely. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Any machine's terminal in a browser tab Shared AI prompt library with project context ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. --- # Control VibeControls from Claude Code with 600+ tools — VibeControls Use Cases > Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. Source: https://vibecontrols.com/use-cases/mcp-server-claude ## Control VibeControls from Claude Code with 600+ tools Ask Claude to start sessions, run actions, manage vibes — all via MCP AI-native development is about more than writing code faster — it's about letting AI manage the environment while you focus on the problem. VibeControls' MCP server makes your entire dev environment controllable by AI. Add the VibeControls MCP server to your Claude Code settings with a one-line config. Authenticate via device-code flow once. Claude now has access to 600+ tools across 22 modules: start or stop sessions, list and create vibes, run VibeDeck tiles, create and delete tunnels, manage agents, query the audit log, trigger scheduled jobs, and more. Ask Claude to "start a tmux session named dev on my local agent and run npm install" — it does it and streams the output back. Ask it to "forward port 8080 to a public URL and give me the link" — it creates the tunnel and returns the URL. Ask it to "run the integration tests VibeDeck tile and show me the output" — it executes the tile and pastes the result. VibeControls MCP also exposes 7 resources (your workspace inventory, agent status, session list) and 7 prompts (common operational query templates) for context. Claude can reason about your environment state, not just take blind actions. For teams using Claude Code at scale, this means the AI agent that writes your code can also manage the environment that runs it. ### Do it yourself Connect the VibeControls MCP server to Claude Code so Claude can drive your sessions, tunnels, vibes, and VibeDeck tiles through natural-language commands. - 1 Add the VibeControls MCP server to Claude Code Claude Code Open Claude Code's MCP configuration and add the VibeControls MCP server with a single one-line entry. Save the config and restart Claude Code so it picks up the new server. @@MD@@``` claude mcp add vibecontrols -- npx -y @vibecontrols/mcp ```@@MD@@ You should see: Claude Code boots the vibecontrols-mcp server and shows its full catalog of tools, resources, and prompts as available. Mark done - 2 Sign in with device-code login Claude Code Ask Claude to log in to VibeControls. It returns a short device code and a link; open the link, enter the code, and pick the workspace you want Claude to act in. - 3 Start a tmux session by asking Claude Claude Code Tell Claude in plain language, for example "Start a tmux session and run npm install." Claude calls the matching VibeControls session tool for you. - 4 Expose a port to a public URL Claude Code Ask Claude to "forward port 8080 to a public URL." Claude creates the tunnel through VibeControls on your behalf. - 5 Run a VibeDeck tile from chat Claude Code Ask Claude to "run the integration-tests VibeDeck tile." Claude triggers the tile and waits for it to finish. - 6 Let Claude pull in live context Claude Code Ask any follow-up like "What sessions are running?" or "Summarize my open tunnels." Claude draws on the built-in VibeControls resources and prompts to ground its answers. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Fine-grained RBAC without complexity Build a custom Vibe plugin by describing it in plain language ### Related use cases AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Review AI-generated scripts in an isolated sandbox before real execution Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. AI-Native & Extensibility ### Give external beta testers time-limited API access via tunnel token Provide beta testers with time-limited API access. Create a public-access tunnel with a 4-hour expiry for testing. --- # Quality scorecards your CI keeps honest — VibeControls Use Cases > Create scorecards for vibes and components. Metrics update with each build, showing pass/warn/fail status and trend history. Source: https://vibecontrols.com/use-cases/metric-scorecards-ci ## Quality scorecards your CI keeps honest Weighted 0–100 scores, CI-pushed metrics, trend sparklines, worst-first portfolio Weekly quality reviews tend to run on anecdotes: whoever speaks loudest sets the agenda, while the coverage numbers that should drive the conversation sit unread in CI logs. "How healthy is this service, really?" deserves a number, not an opinion. VibeControls metric scorecards give every vibe, catalog, and component a weighted 0–100 score built from individual metrics — each with its own threshold, weight, and direction (higher-is-better for coverage, lower-is-better for open vulnerabilities or build time). Apply a scorecard template — built-in starters are seeded into every workspace — and a production-readiness check set lands on the entity in one click. Then connect CI. Rotate a workspace push token (shown once, stored hashed — the raw value never persists), add it to your pipeline, and have every build POST its metrics to the scorecard push endpoint. Coverage, test counts, lint warnings — whatever your pipeline measures becomes a scorecard metric with a recorded history point per push. Status is derived server-side from the threshold; a pipeline can report a value, but it can never assert its own pass. When a metric dips below its threshold, the card flips to failing automatically — with the remediation guidance you attached, so the person looking at it knows what to do next, not just that something is red. History powers trend sparklines per metric, and each metric can embed a Grafana panel for the full picture on demand. The payoff is the portfolio panel: every scored entity in a catalog, sorted lowest score first, trends alongside. The weekly review now starts from "worst score first" — and the argument about which service needs attention is over before it starts. ### Do it yourself Set up a weighted 0–100 quality scorecard on any vibe, catalog, or component, wire your CI pipeline to push build metrics, and watch trends and a worst-first portfolio keep your whole workspace honest. - 1 Open the Scorecard tab /vibecontrols/vibes/:id Open the vibe you want to score and click the Scorecard tab. (Catalogs and components have the same tab — open one at /vibecontrols/catalogs/:id or /vibecontrols/components/:id to score it instead.) You should see: The Scorecard panel opens with a friendly empty state inviting you to apply a template or push from CI. [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Apply a ready-made template /vibecontrols/vibes/:id Under "Apply a template", click a built-in scorecard like "Service Production-Readiness" or "AI Agent Quality". - 3 Generate your CI push token /vibecontrols/vibes/:id Under "Push from CI", click Generate token, then copy the raw token immediately. - 4 Wire the push endpoint into your pipeline Terminal Copy the provided curl snippet and add a step to your CI pipeline that posts your build's metrics to the scorecard push endpoint, using your token as a Bearer authorization header. - 5 Reload to see the live score /vibecontrols/vibes/:id Refresh the Scorecard tab after your build runs. - 6 Expand a metric for trends and dashboards /vibecontrols/vibes/:id Click a metric card to expand it. To embed a live dashboard, add a metric with an Embed URL (for example, a Grafana panel link). - 7 Compare your whole workspace /vibecontrols/vibes/:id Scroll to "Compare across the workspace" at the bottom of the Scorecard tab. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Pair on a live terminal session from a Slack link Full audit trail for every operation, every actor, every outcome ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. --- # New hire productive in under an hour — VibeControls Use Cases > Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Source: https://vibecontrols.com/use-cases/new-hire-onboarding ## New hire productive in under an hour Clone, bind, go — all dev context lives in the Vibe First-week developer onboarding typically involves: hunting down the .env, figuring out which tmux session to attach to, reading a 300-line setup wiki, asking three people for credentials, and hoping the wiki was updated after the last infrastructure change. When your projects are Vibes in VibeControls, that story is completely different. The Vibe carries the environment variables (encrypted), the documentation, the shared sessions, the VibeDeck for common operations, and the AI context for the codebase. The new hire installs the CLI, runs 'vibe start', and opens VS Code. The VibeControls extension shows them the team Vibe in the sidebar. They bind their local project folder to it. Done. The environment variables flow in. The documentation is embedded in the Vibe detail view. The team's active sessions are visible. The VibeDeck with "Run tests", "Start dev server", and "Check DB" is ready to use. The AI context is attached to prompts. The onboarding ceremony is now: install CLI, bind Vibe, start coding. The wiki doesn't drift because the source of truth is the Vibe, not a document. ### Do it yourself A new hire installs the CLI, registers their laptop, binds their local project folder to the existing team Vibe in VS Code, and instantly gets the team's environment variables, docs, sessions, VibeDeck actions, and AI context. - 1 Install the VibeControls CLI Terminal Open your terminal and install the VibeControls CLI globally with npm. @@MD@@``` npm install -g @vibecontrols/cli ```@@MD@@ You should see: The vibe command is now available on your machine. Mark done - 2 Register your laptop as a live agent Terminal Run vibe start to install the agent and bring your laptop online. - 3 Open VS Code and find the team Vibe VS Code Open your cloned project in VS Code and click the VibeControls icon to open the Vibes sidebar. - 4 Bind your folder to the team Vibe VS Code Click the bind button on the Vibes view (or run "Bind Folder to Existing Vibe" from the command palette) and pick the team Vibe from the list. - 5 Read the embedded team docs /vibecontrols/vibes/:id/docs Open the team Vibe in the web app and click the Docs tab to browse the team's onboarding documentation. - 6 See the team's shared sessions /vibecontrols/sessions Open the Sessions view to see the live sessions your team has shared with you. - 7 Run a ready-made VibeDeck action /vibecontrols/vibedeck Open the VibeDeck and click a tile like "Run tests", "Start dev server", or "Check DB". - 8 Ask the AI with project context attached VS Code Use the VibeControls "Send to AI" command in VS Code to ask a question about the codebase. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams The entire platform from your terminal — and your CI pipeline Golden-path templates that preserve service conventions ### Related use cases Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Platform team's shared ops deck eliminates 80% of tickets Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. Team & Platform ### Living software catalog with real ownership and dependency graphs Maintain a live software catalog with VibeControls. Register services, APIs, and libraries with ownership and dependency graphs.'s connected to the Vibes where work actually happens. --- # Queue AI test-generation jobs and review results in the morning — VibeControls Use Cases > Queue AI test jobs overnight with VibeControls. Review generated test cases in the morning, ready to integrate into your test suite. Source: https://vibecontrols.com/use-cases/nightly-ai-test-queue ## Queue AI test-generation jobs and review results in the morning Let AI write test cases overnight — you review and approve Test coverage for a growing API surface is a perpetual backlog item. Writing test cases is important but slow, and AI can generate a credible first draft faster than any engineer — but only if you give it time and context. VibeControls AI Queue makes this a background workflow. Create a prompt template: "For each of these API endpoints, generate comprehensive unit tests covering happy path, validation errors, and edge cases: {{endpoints}}". Attach your API schema file as an AI Context. Before you leave for the day, run the prompt with Queue — not Dispatch. VibeControls logs it as a pending task. Overnight, your AI provider executes it and stores the result in the queue. You open the queue in the morning and see the completed runs. Each run shows which model ran it, how many tokens it consumed, and the full response. The generated test cases are ready to review, edit, and copy into your test files. This turns AI from an interactive tool into an async collaborator. You set the agenda, the AI works overnight, and you review and approve. ### Do it yourself Queue AI test-generation prompts to run overnight against your API schema, then review each completed run's model, token usage, and full response in the morning. - 1 Create your test-generation prompt /vibecontrols/ai-tools/ai/prompts Open the AI Prompts tab and create a new prompt template describing the tests you want generated. Save it. You should see: Your test-generation prompt is saved and ready to reuse. [Open in app](https://app.vibecontrols.com/vibecontrols/ai-tools/ai/prompts) Mark done - 2 Attach your API-schema context /vibecontrols/ai-tools/ai/contexts Switch to the Contexts tab and create an AI Context holding your API schema, then save it so the prompt has the API details to work from. - 3 Queue the prompt to run overnight /vibecontrols/ai-tools/ai/prompts Open your prompt, pick the API-schema context, and click Queue instead of Dispatch. This schedules the job to run asynchronously rather than right now. - 4 Let it run while you sleep /vibecontrols/ai-tools/ai/prompts Leave the Queue tab open. Your AI provider works through the queued jobs overnight while the tab auto-refreshes on its own. - 5 Review the completed runs in the morning /vibecontrols/ai-tools/ai/prompts Come back to the Queue tab and scan the list of finished jobs. Click any completed item to open its run. - 6 Copy the generated tests into your project /vibecontrols/ai-tools/ai/prompts Open a completed run, read through the generated tests, and copy the response straight into your test files. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Instant Slack alert when a session fails Roll out new capabilities to your whole fleet from the plugin catalog ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Enterprise SSO via OIDC — every engineer logs in with their corporate identity — VibeControls Use Cases > Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Source: https://vibecontrols.com/use-cases/oidc-sso-enterprise ## Enterprise SSO via OIDC — every engineer logs in with their corporate identity Okta, Microsoft Entra ID, Google Workspace, or any OIDC provider — zero password sprawl Managing developer tooling credentials separately from corporate identity is a compliance liability. When a developer leaves, the IT checklist has 30 line items — Jira, GitHub, Slack, CI/CD, staging databases, and a long tail of "miscellaneous tools". VibeControls should not be on that list. SSO is configured once at the platform/tenant level by your identity administrator — connect your provider (Okta, Microsoft Entra ID, Google Workspace, or any compliant OIDC provider) and it governs VibeControls login alongside the rest of the platform. Developers log in via the corporate login button, are authenticated against your IdP, and their VibeControls identity is tied to their corporate account. When you deprovision the engineer in your IdP, their VibeControls session tokens are revoked on next refresh. No manual de-provisioning step, no window of continued access. For the CLI, developers use OAuth device-code flow to authenticate without password prompts. For service-to-service automation, API keys are issued separately and are visible in the workspace audit trail. Combined with RBAC (Owner, Admin, Member, Viewer, and Guest workspace roles), VibeControls gives enterprise IT teams the credential governance and lifecycle management they require without compromising developer velocity. ### Do it yourself Connect your corporate identity provider so every engineer signs in to VibeControls with their company login, with CLI device-code auth and API keys for automation. - 1 Open SSO Configuration Admin Console As your organization's admin, open the workspace Admin Console and go to SSO Configuration. You'll see tabs for SAML, OIDC, Clients, and Test. You should see: The SSO Configuration page opens with the SAML, OIDC, Clients, and Test tabs ready to set up. Mark done - 2 Connect your identity provider Admin Console On the OIDC (or SAML) tab, paste your provider's metadata URL from Okta, Microsoft Entra ID, or Google Workspace, then save. Use the Test tab to run a verification login. - 3 Sign in with corporate identity /auth/login Engineers open the VibeControls login page and choose the single sign-on option. They're redirected to your corporate identity provider to authenticate. - 4 Log in from the CLI Terminal On a developer machine, run the CLI login command. It prints a device-code URL to open and approve in the browser. - 5 Issue API keys for automation /devportal For service automation, open the Developer Portal and mint an API key for your integrations. Revoke any key at any time to instantly cut its access. - 6 Deprovision through your IdP Identity Provider When someone leaves, disable their account in your identity provider as usual — no extra step needed in VibeControls. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs Full audit trail for every operation, every actor, every outcome Security scorecards that show which services are production-ready ### Related use cases Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. Enterprise & Governance ### Connect AI coding tools to an agent-local plan workspace Set up the agent-local Plannotator integration for supported AI CLIs, then use its embedded workspace to inspect the plan sessions those tools create. --- # Connect AI coding tools to an agent-local plan workspace — VibeControls Use Cases > Set up the agent-local Plannotator integration for supported AI CLIs, then use its embedded workspace to inspect the plan sessions those tools create. Source: https://vibecontrols.com/use-cases/plan-mode-ai-safety ## Connect AI coding tools to an agent-local plan workspace Install Plannotator, configure a supported CLI hook, and review plans in its embedded workspace AI coding tools can benefit from a deliberate planning surface, but the useful boundary is clear: VibeControls connects a registered agent to the plan provider; the configured AI CLI and provider control when and how a plan is produced. On an online agent, install the Plan orchestrator and Plannotator provider. The Plan workspace can install the local Plannotator binary, detect supported Claude Code, OpenCode, Codex, Pi, or Gemini CLIs, and configure that CLI's hook. Once configured, VibeControls embeds the provider workspace through the agent tunnel so you can inspect its active plan sessions without separately exposing the agent. Use the provider workspace to review a plan and its annotations according to the provider and CLI you configured. VibeControls does not currently provide a workspace-level blast-radius threshold, automatic approval ledger, or idle-plan watchdog; those execution and approval semantics remain with the selected CLI and plan provider. ### Do it yourself Connect a supported AI CLI on your agent to the Plannotator plan workspace, then inspect the plan sessions the configured provider creates. - 1 Install the plan plugin on your agent /vibecontrols/plugins On a healthy, online agent, open the Plugins page and install the plan plugin. You can also run the install command from your terminal. You should see: The plan provider (Plannotator) becomes available to your agent and appears as a selectable provider. [Open in app](https://app.vibecontrols.com/vibecontrols/plugins) Mark done - 2 Install the plan plugin from the terminal Terminal If you prefer the command line, install the plan plugin directly on the machine running your agent. - 3 Open Plan mode /vibecontrols/ai-tools/plan Go to AI Tools and switch to the Plan tab to open the full-screen Plannotator review panel. - 4 Connect your AI CLI /vibecontrols/ai-tools/plan Let the Plan panel detect your AI CLI (Claude Code, OpenCode, Codex, Pi, or Gemini). If it's not yet hooked up, follow the on-screen setup instructions to enable the plan hook. - 5 Dispatch a task in Plan mode /vibecontrols/ai-tools/plan Run a task with the supported CLI whose hook you configured. When that CLI invokes its plan hook, return to the Plan panel to inspect the provider session. - 6 Review, annotate, and approve /vibecontrols/ai-tools/plan In the embedded Plannotator portal, read the plan and use the provider’s available review and annotation controls. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs Security scorecards that show which services are production-ready Fine-grained RBAC without complexity ### Related use cases Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. --- # Roll out new capabilities to your whole fleet from the plugin catalog — VibeControls Use Cases > Manage plugins across your fleet with VibeControls. Browse, install, and update plugins without SSH, directly from the catalog. Source: https://vibecontrols.com/use-cases/plugin-catalog-management ## Roll out new capabilities to your whole fleet from the plugin catalog Browse, install, update, remove — per agent, prerequisites handled, no SSH Rolling out a new tool to a fleet of machines is the classic DevOps grind: SSH into each box, install the binary, chase down missing dependencies, and hope every server ends up on the same version. Six months later, half the fleet has drifted. VibeControls makes capabilities installable. The plugin catalog organizes everything in three tiers: meta plugins define a capability interface (session-manager, tunnel, ai, storage, plan, security), provider plugins implement it (tmux, Cloudflare, Claude, Plannotator), and standalone tool plugins add features outright (ssh, git, graphiql, code-server, backup). All of it distributes via npm as @vibecontrols/vibe-plugin-* packages. Installation is one click from the catalog page — pick the plugin, pick the agent. The agent runs a prerequisite check first: if a dependency is missing, you're shown exactly what needs installing and asked to approve it (sudo is prompted only when actually required). Once loaded, the plugin mounts its REST routes and UI surfaces, and the per-agent inventory shows it healthy with its version. The rest of the lifecycle is just as governed. Update pulls the new npm version and the agent reloads the plugin. Remove unmounts it cleanly — meta plugins that depended on it degrade gracefully instead of breaking sessions mid-flight. And a failed prerequisite never leaves a half-loaded plugin masquerading as healthy. Want the Cloudflare tunnel provider on three staging agents? Three clicks from the catalog page, watch each inventory go green, done. The team gets tunnels and nobody opened a single SSH connection. And when the catalog doesn't have what you need, AI Builder can author a custom plugin — but for everything that already exists, install beats build. ### Do it yourself Roll out a new capability to your whole fleet by installing a plugin from the catalog onto your agents, approving any prerequisites, and watching each agent's inventory go green. - 1 Browse the plugin catalog /vibecontrols/plugins Open the Plugins page and explore the catalog, organized into three tiers: meta plugins, provider plugins, and standalone tools. Use the search box or category filters to find the capability you want to roll out. You should see: You see a searchable grid of plugins grouped by tier, each card showing what the plugin does and its npm package. [Open in app](https://app.vibecontrols.com/vibecontrols/plugins) Mark done - 2 Install it on an agent from the catalog /vibecontrols/plugins On the plugin you want, click Install on agent and pick a healthy agent from the picker. Confirm to start the install. - 3 Approve the prerequisite check /vibecontrols/plugins If the plugin needs a dependency, the agent runs a prerequisite check and shows you exactly what's missing. Review the list and click Approve; enter your password only if a step actually requires elevated access. - 4 Confirm it's healthy with its version /vibecontrols/agents/:id Open the agent's detail page and switch to the Plugins tab, Installed list. Check that the new plugin appears. - 5 Roll it out to the rest of your fleet /vibecontrols/plugins Back on the catalog, select your remaining staging agents and install the same plugin on each. Approve any prerequisites as they come up. - 6 Update a plugin when a new version ships /vibecontrols/agents/:id On an agent's Plugins tab, Installed list, click Update next to the plugin and choose the version you want. - 7 Remove a plugin cleanly /vibecontrols/agents/:id On the agent's Installed plugins list, click Remove next to the plugin and confirm. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Queue AI test-generation jobs and review results in the morning The entire platform from your terminal — and your CI pipeline ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Fine-grained RBAC without complexity — VibeControls Use Cases > Define workspace roles with RBAC. Roles map to access needs, providing project-level control without added complexity. Source: https://vibecontrols.com/use-cases/rbac-workspace-roles ## Fine-grained RBAC without complexity Owner, Admin, Member, Viewer, Guest — with per-resource ownership that just works Access control systems are usually either too coarse (everyone is admin) or too complex (200-line policy files nobody understands). VibeControls takes a third path: a small set of workspace roles — Owner, Admin, Member, Viewer, and Guest — with a simple ownership model that matches how engineering teams actually work. Viewers can see everything — agents, sessions, vibes, operations — but can't change anything. Good for stakeholders, auditors, or new team members who need visibility before getting permissions. Guests get an even narrower, scoped view for external collaborators. Members can create and manage the resources they own: their own agents, vibes, sessions, configurations, notes. They can't touch resources owned by others unless those resources are explicitly shared. This is the right model for a team where everyone has a lane. Owners and Admins have full control over all workspace resources — typically one or two people per workspace, with the Owner sitting at the top of the chain. Admin operations are all captured in the audit log so "with great power comes great accountability" is enforced by the system. Per-vibe visibility adds a layer on top: resources attached to a Vibe can be visible only to the owner (PRIVATE) or to anyone with Vibe access (SHARED_WITH_VIBE). Teams that work on shared projects get shared visibility without everyone needing Admin. Invite a member, set their role, and they inherit the right permissions immediately. Remove a member and their access is revoked on the next gateway check — every request is re-evaluated, so a removed member is locked out the moment they next touch the platform. ### Do it yourself Invite a teammate, assign their workspace role, control per-vibe visibility, and review every admin action so the right people get exactly the right access. - 1 Invite a teammate to the workspace /workspaces Open Workspaces from the app navigation and use the invite control to send an invitation to your teammate's email. You should see: The person receives an invitation and joins your workspace as a member. [Open in app](https://app.vibecontrols.com/workspaces) Mark done - 2 Assign their role /rbac Open RBAC from the app navigation, find the new member, and pick a role such as Workspace Admin, Member, Viewer, or Guest. - 3 See permissions applied automatically /vibecontrols/audit Have the member browse any VibeControls page. They can view shared resources, but action buttons they aren't allowed to use are hidden or disabled. - 4 Control who can see each linked entity /vibecontrols On a vibe-linked entity, open the link or share control and set its visibility to Private or Shared with the vibe. - 5 Review who did what /vibecontrols/audit Open the Audit page to see a chronological trail of administrative actions across the workspace. - 6 Remove a member to revoke access /workspaces Go back to Workspaces and remove the member you want to off-board. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs Connect AI coding tools to an agent-local plan workspace Control VibeControls from Claude Code with 600+ tools ### Related use cases Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. --- # Bootstrap a multi-server agent fleet in under 5 minutes — VibeControls Use Cases > Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. Source: https://vibecontrols.com/use-cases/remote-fleet-bootstrap ## Bootstrap a multi-server agent fleet in under 5 minutes SSH targets → registered agents, no manual binary uploads Standing up a new environment used to mean SSHing into each server, downloading the right binary, writing a systemd unit, configuring credentials, and manually verifying connectivity. With ten servers, that's an afternoon. VibeControls cuts this to minutes. Stand up your first node locally with "vibe start" — that running agent drives the remote installs. Then register each server as an SSH Target: hostname, port, credentials. VibeControls validates connectivity and tests the SSH handshake. From there, click "Install Agent" on any registered target and the platform pushes the agent over your source agent's SSH connection, writes the autostart configuration, and registers the agent with your workspace. You watch each agent come online in real time — health status, version, platform info. Once online, you can start sessions, install plugins, configure Vibes, and run actions from the same web UI you use for your local machine. No Ansible playbooks, no manual SSH, no agent version drift across servers. Every machine in your fleet is under the same governance model — RBAC, audit, plugin management — from day one. ### Do it yourself Onboard your first server locally, then register every other server as an SSH target and push the VibeControls agent to each one from the dashboard, watching your whole fleet come online in real time. - 1 Start your first agent Terminal On any one machine you control, install and start the VibeControls agent. This first agent becomes the launchpad that pushes the agent out to every other server for you. @@MD@@``` npm install -g @vibecontrols/agent && vibe start ```@@MD@@ You should see: Your first agent registers with your workspace and appears online, ready to drive remote installs. Mark done - 2 Register a server as an SSH target /vibecontrols/agents?tab=targets Open the Targets tab and click Create Target. Choose SSH, then enter the server's hostname, port, and credentials (password or private key). - 3 Test the connection /vibecontrols/agents?tab=targets On the target card, click Test. VibeControls opens an SSH handshake to confirm the credentials and reachability. - 4 Install the agent /vibecontrols/agents?tab=targets Open the target card's menu and click Install Agent. VibeControls connects over SSH, sets up the agent, configures it to start automatically, and registers it with your workspace. - 5 Watch it come online in real time /vibecontrols/agents?tab=targets Stay on the Targets tab and watch the install job. Each step updates live as the agent is set up. - 6 See the new agent in your fleet /vibecontrols/agents Open the Agents page to find the freshly installed agent listed alongside the rest. - 7 Repeat for every server /vibecontrols/agents?tab=targets Add an SSH target and run Install Agent for each remaining server. From there you can attach plugins and start sessions on any of them. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Give your AI assistant scoped access to your codebase Visual map of your entire infrastructure topology ### Related use cases DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. DevOps / SRE ### One-click incident response operations deck Streamline incident response with VibeDeck. Execute common actions like API restarts and cache flushes with a single click. --- # Review AI-generated scripts in an isolated sandbox before real execution — VibeControls Use Cases > Run AI-generated scripts in a VibeControls Sandbox. Review output in isolation before executing in a real environment. Source: https://vibecontrols.com/use-cases/sandbox-safe-execution ## Review AI-generated scripts in an isolated sandbox before real execution AI writes the deploy script — sandbox runs it — you approve before touching prod AI-generated scripts are usually mostly right and occasionally catastrophically wrong. The problem is you can't tell which it is without running it — and running it on a real server means committing to the outcome. VibeControls Sandboxes are designed for exactly this scenario. When an AI session generates a deploy script, a migration, or a configuration change, instead of running it directly, send it to a Sandbox. A Sandbox is a configured, isolated execution environment — Cloudflare Worker or custom compute — with specific resource limits, network egress rules, and secret bindings. It can't accidentally reach your production database or write to your production S3 bucket unless you explicitly configure it to. Send the AI-generated script to the Sandbox. It runs, captures stdout/stderr, exits. You see the output in the Sandbox logs: was the migration valid? Did the script produce the expected output? Are there any error messages that suggest the script needs revision? Only when you're satisfied that the sandbox run looks correct do you execute against the real target. The sandbox output is logged to the workspace audit trail alongside the eventual real execution — so you have both the test run and the production run in the same record. AI acceleration without AI recklessness. ### Do it yourself Run an AI-generated script in an isolated VibeControls sandbox, review its captured output, and only then promote it to a real execution — all in one audited record. - 1 Generate a script with the AI assistant /vibecontrols Open the AI assistant panel and ask it to produce your deploy script, migration, or config change. Copy the script it generates. You should see: The assistant returns a ready-to-test script that you can send to a sandbox instead of running it directly. [Open in app](https://app.vibecontrols.com/vibecontrols) Mark done - 2 Create an isolated sandbox /sandbox/new Go to the new-sandbox page, pick a template or image, and create the sandbox. It runs with resource limits, egress rules, and secret bindings already applied. - 3 Open your sandbox once it's running /sandbox/:id Open the sandbox detail page and wait until its status shows Running. - 4 Run the script in the Script Runner /sandbox/:id In the Script Runner card, choose an interpreter (bash, sh, python, or node), paste your script, and click Run. - 5 Review the captured output /sandbox/:id Open the run in the history list to inspect its captured stdout, stderr, and exit code side by side. - 6 Promote the reviewed run to the real target /sandbox/:id Once you're satisfied with a successful run, click Promote on that run to sign it off for the real target. Ready to make this your story? ### Capabilities used ### Category For developers building AI-first workflows and custom tooling Build a custom Vibe plugin by describing it in plain language Give external beta testers time-limited API access via tunnel token ### Related use cases AI-Native & Extensibility ### Control VibeControls from Claude Code with 600+ tools Integrate VibeControls MCP server with Claude Code. Access 600+ tools to manage your development environment naturally. AI-Native & Extensibility ### Build a custom Vibe plugin by describing it in plain language Describe your plugin in plain text. VibeControls AI Builder creates a complete plugin with UI panels and test environment. AI-Native & Extensibility ### Give external beta testers time-limited API access via tunnel token Provide beta testers with time-limited API access. Create a public-access tunnel with a 4-hour expiry for testing. --- # Centrally review scheduled database-backup runs across servers — VibeControls Use Cases > Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup Source: https://vibecontrols.com/use-cases/scheduled-db-backup ## Centrally review scheduled database-backup runs across servers One calendar view, every backup, every server — with execution history and logs The hardest part of database backups isn't running them — it's knowing they ran. In a distributed environment, backups are often cron jobs scattered across different servers with no central visibility. You find out they stopped working when you need to restore. VibeControls Vibe Calendar changes the model. Attach your database action to a schedule — every night at 02:00, run the backup script. VibeControls dispatches the job to the selected agent, captures execution status and output, and keeps a run history for the scheduled job. The Calendar dashboard shows every scheduled job across the workspace in a single view. Open a job to inspect its executions, status, duration, and available output. If a job fails, you can drill into the recorded run instead of starting with SSH. The scheduler can run a deliberately configured webhook job, but the VibeControls Webhooks page does not currently offer an automatic “scheduled backup failed” subscription. “Did the backup run?” is a calendar and execution-history check today. ### Do it yourself Schedule a nightly database backup script on a selected server, then review its execution history and available output from the Calendar dashboard. - 1 Define your backup Action /vibecontrols/actions Click Create and define a new Action that runs your backup script or command, then choose the agent on the database server that should run it. You should see: A reusable backup Action is saved and bound to the agent on that server, ready to be scheduled. [Open in app](https://app.vibecontrols.com/vibecontrols/actions) Mark done - 2 Open the Calendar /vibecontrols/calendar Go to the Calendar to see your month, week, day, and agenda views of scheduled work. - 3 Schedule the backup nightly at 02:00 /vibecontrols/calendar Click Schedule, pick the CRON option and set it to run every night at 02:00, then link your backup Action. - 4 Let it run and capture each result /vibecontrols/calendar Leave the schedule running — each night VibeControls automatically dispatches the backup to the correct server's agent and records the output. - 5 See every server's jobs in one view /vibecontrols/calendar Use the Calendar's agenda view to scan all your scheduled backups, grouped and filterable by server, in a single place. - 6 Drill into a run's log /vibecontrols/calendar Click any scheduled job to open its detail page, then review the list of runs and open a run to read its full output, exit code, and status. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Visual map of your entire infrastructure topology One-click incident response operations deck ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### One-click incident response operations deck Streamline incident response with VibeDeck. Execute common actions like API restarts and cache flushes with a single click. --- # Scan every vibe and agent — then triage the findings in one queue — VibeControls Use Cases > Conduct comprehensive security scans on Vibes and agents. Manage findings in a unified queue, set statuses, and document exceptions efficiently. Source: https://vibecontrols.com/use-cases/security-scan-findings ## Scan every vibe and agent — then triage the findings in one queue Run staged security scans, catch risky SSH config, and work findings to resolution with statuses and documented exceptions Production-readiness tiers tell you where a service stands; they do not run the scan. Security teams need the active workflow underneath: kick off a real scan, collect the findings, and drive each one to a decision. VibeControls builds that workflow into the platform. Open the Security tab on a Vibe and run a scan for the lifecycle stage that matters — DEVELOPER_LOCAL while someone is hacking on a branch, PULL_REQUEST_FAST on a review, BUILD or MAIN_MERGE in the pipeline, PROMOTE_PROD before a release. The scan produces a run with a status and conclusion, and a list of findings categorized by type: VULN, SECRET, SAST, CONFIG, LICENSE, POLICY, or SUPPLYCHAIN. Nothing is buried in a CI log you have to go dig for — the run and its findings live in the workspace. Agents get scanned too. Point VibeControls at an agent and scan its SSH configuration; it enumerates the configured hosts so a machine quietly trusting a host it should not becomes visible before it becomes an incident. Then comes the part that actually closes the loop: triage. Every finding starts OPEN. Work it and set its status — FIXED once it is patched, ACCEPTED when the risk is understood and owned, FALSE_POSITIVE when the scanner is wrong, SUPPRESSED when it is noise. When you accept a risk, file a documented security exception so the decision is recorded, attributable, and reviewable later — not a Slack message someone half-remembers. A per-repository security config lets you tune how a Vibe is scanned. Because scans and findings are workspace resources governed by RBAC and captured in the audit trail, the whole program is auditable: you can show which scans ran, what they found, who triaged each finding, and why any exception exists. That turns ad-hoc scanning into a security posture you can actually stand behind. ### Do it yourself Run a staged security scan against a Vibe, scan an agent’s SSH config, and triage every finding to resolution with statuses and documented exceptions — all from the Security tab. - 1 Open the Security tab on your Vibe /vibecontrols/vibes Open the project Vibe and switch to the Security tab. You should see: You see scan history and the current findings for this Vibe. [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Run a scan for the right lifecycle stage /vibecontrols/vibes Click Run scan and pick the stage — Developer Local, Pull Request, Build, Main Merge, or Promote to Prod. - 3 Scan an agent’s SSH configuration /vibecontrols/agents Open an agent, go to its Security tab, and scan its SSH config. - 4 Triage each finding /vibecontrols/vibes Work each finding and set its status — Fixed, Accepted, False Positive, or Suppressed. - 5 Document an exception when you accept a risk /vibecontrols/vibes When you accept a finding, file a security exception recording the rationale. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs See exactly what your AI agents did — task by task Package any operation into a reusable, shareable action ### Related use cases Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Security scorecards that show which services are production-ready Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation.'t make the bar are flagged before they reach production. --- # Security scorecards that show which services are production-ready — VibeControls Use Cases > Grade services with security scorecards. Checklists include ownership, documentation, secrets scan, and SBOM generation. Source: https://vibecontrols.com/use-cases/security-scorecard ## Security scorecards that show which services are production-ready Bronze / Silver / Gold tiers — owners, docs, security checks, monitoring "Is this service ready for production?" is a question that deserves a systematic answer, not an opinion. VibeControls security scorecards make production-readiness a measurable, tracked property of every service. The scorecard system runs weighted checks against services registered in the catalog. Checks are organized into tiers: Bronze (baseline — has an owner, has a README, passes a secrets scan), Silver (intermediate — SBOM generated, DAST clean, no critical CVEs), Gold (production — all Silver checks, plus runtime security monitoring, incident runbook documented). Checks are populated by the security plugin ecosystem. Secrets scan results come from vibe-plugin-security-secrets-pr. SBOM build results from vibe-plugin-security-sbom-build. Custom policy rules run through the release-gate plugin. The scorecard aggregates them. The catalog view shows each service's current tier. Red badges on services that have regressed. Trend lines show whether security posture is improving or declining. Security officers can filter by tier to find the services that most need attention — and link directly to the failing checks for a remediation plan. Release gates can block a deployment if a service hasn't met its tier requirements. Scorecards go from vanity metric to deployment prerequisite. ### Do it yourself Grade each catalog service against a Bronze/Silver/Gold security checklist, feed real scan results into the scorecard from CI, and see each service's production-readiness tier and remediation right in the catalog. - 1 Open a catalog service /vibecontrols/catalogs/:id Go to your service catalog and click into the service you want to grade for production-readiness. You should see: The service detail page opens with tabs across the top, including a Scorecard tab. [Open in app](https://app.vibecontrols.com/vibecontrols/catalogs) Mark done - 2 Open the Scorecard tab /vibecontrols/catalogs/:id Click the Scorecard tab and choose the built-in "Security Production-Readiness" template. - 3 Run a security scan on the vibe /vibecontrols/vibes/:id Open the linked vibe, go to its Security tab, and run a scan to check secrets, SBOM, CVEs, and runtime readiness. - 4 Push results from CI into the scorecard Terminal From your CI pipeline, have the security plugins post their scan results to the scorecard using your push token, so every check is backed by a real run. - 5 See the production-readiness tier /vibecontrols/catalogs/:id Return to the Scorecard tab and read the tier badge on the header. - 6 Fix failing checks to climb tiers /vibecontrols/catalogs/:id Click any failing check to see what's missing — like a README, an SBOM, or no critical CVEs — and address it, then re-push from CI. Ready to make this your story? ### Capabilities used ### Category For security officers, IT admins, and compliance-focused orgs Enterprise SSO via OIDC — every engineer logs in with their corporate identity Connect AI coding tools to an agent-local plan workspace ### Related use cases Enterprise & Governance ### Full audit trail for every operation, every actor, every outcome Capture operations in a tamper-evident audit log with actor identity and timestamp. Logs are retained for 7 years. Enterprise & Governance ### Enterprise SSO via OIDC — every engineer logs in with their corporate identity Set up OIDC SSO for VibeControls. Authenticate users via corporate identity provider, ensuring secure access management. Enterprise & Governance ### Connect AI coding tools to an agent-local plan workspace Set up the agent-local Plannotator integration for supported AI CLIs, then use its embedded workspace to inspect the plan sessions those tools create. --- # Self-serve operations for the whole org via shared VibeDeck — VibeControls Use Cases > Enable self-serve operations with shared VibeDeck. Product teams can deploy, flush caches, and perform health checks without SSH access. Source: https://vibecontrols.com/use-cases/share-vibedeck-org ## Self-serve operations for the whole org via shared VibeDeck Platform team publishes a deck — everyone executes without needing access Platform teams spend a surprising fraction of their time fielding requests that are inherently self-service: "can you restart the staging API?", "can you flush the cache?", "can you check the disk on the worker?". These are mechanical tasks that shouldn't require platform team involvement — but they do, because the only alternative is giving everyone SSH keys. VibeControls VibeDeck changes the access model. Build the deck with the safe, pre-approved operations. Set each tile to run on a specific agent. Then share the deck: enable "Public access", optionally require a simple token, and toggle "Allow execution". The link goes into your internal tools wiki or Slack channel description. Now any engineer can click "Restart staging API" from their browser without SSH credentials, without paging anyone, and with full audit trail — the execution is logged with actor identity, timestamp, and output. The platform team defines the operations once and reviews the audit log periodically. Everyone else self-serves. The "can you just restart it?" message disappears. ### Do it yourself Build a VibeDeck of safe operations tiles and share it as a public, executable link so any engineer in your org can run trusted ops from their browser without SSH credentials. - 1 Build your operations deck /vibecontrols/vibedeck Open VibeDeck and add tiles for the safe operations you want to share, such as "Restart staging API". Set each tile to run on the specific agent that should execute it. You should see: Your deck shows a grid of one-click action tiles, each bound to the right agent and ready to run. [Open in app](https://app.vibecontrols.com/vibecontrols/vibedeck) Mark done - 2 Open the public link panel /vibecontrols/vibedeck With your deck open, click the "Public link" action to open the public link panel. - 3 Allow execution on the link /vibecontrols/vibedeck Set the access toggle to "Execute" so people opening the link can actually run the tiles, not just view them. - 4 Set expiry, a token, and a usage cap /vibecontrols/vibedeck Choose an expiry window (4 hours, 24 hours, 7 days, or never), optionally add a simple password to require a token, and optionally set a maximum number of uses. - 5 Create and share the link /vibecontrols/vibedeck Click create to mint the link, then copy it and send it to your team. You can revoke it anytime from the same panel. - 6 Engineers run a tile from the browser /vibecontrols/vibedeck Any engineer opens the link (entering the password if you set one) and clicks a tile like "Restart staging API". - 7 Review the audit trail /vibecontrols/vibedeck Each shared run is recorded automatically, so you can review who ran what and when. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers One-click incident response operations deck Connect your database GUI to a remote database over a tunnel ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Per-project environment variables without .env sprawl — VibeControls Use Cases > Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Source: https://vibecontrols.com/use-cases/vibe-env-variables ## Per-project environment variables without .env sprawl Every session on a project inherits the right credentials automatically Managing environment variables across projects is a constant source of "it works on my machine" bugs. A miscopied .env, a stale value, or accidentally reusing a production secret in a dev session — any of these can cost hours. VibeControls solves this at the workspace layer. Create a Vibe for your project and open the Environment tab. Add variables — plain or secret. Secret values are AES-256-GCM encrypted at rest and never shown in plaintext after the first save. When you start a terminal session under that Vibe, the agent injects those variables automatically. Switch to a different Vibe — different variables. Parent Vibes can define shared base config (like a staging database URL) and child Vibes override only what's different. The result: your dev, staging, and production contexts are isolated namespaces, not three identically-named files you have to symlink before every run. It's environment hygiene that requires zero discipline to maintain. ### Do it yourself Store per-project environment variables — plain or encrypted secret — on a Vibe so every terminal session you start under it gets them injected automatically. - 1 Create a Vibe for your project /vibecontrols/vibes Open the Vibes page and click Create. Give the Vibe a name that matches your project. You should see: Your new Vibe appears in the list and opens its detail view. [Open in app](https://app.vibecontrols.com/vibecontrols/vibes) Mark done - 2 Open the Config tab /vibecontrols/vibes From your Vibe's detail page, click the Config tab and find the Environment Variables panel. - 3 Add a plain variable /vibecontrols/vibes Click to add a variable, enter a key and value (for example DB_HOST and db.prod.internal), and leave the Secret toggle off. Save it. - 4 Add a secret variable /vibecontrols/vibes Add another variable, enter its key and value (for example API_KEY), and switch on the Secret toggle before saving. - 5 Confirm your variables persist /vibecontrols/vibes Reload the page and return to the Config tab to review the Environment Variables panel. - 6 Start a session under the Vibe /vibecontrols/vibes Go to the Vibe's Sessions tab and start a new terminal session under this Vibe. - 7 Override with a child Vibe /vibecontrols/vibes Create a child Vibe under this one for a specific environment, then set only the variables you want to change in its Config tab. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Zero-config local agent in under a minute Any machine's terminal in a browser tab ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. Individual Developer ### Instant public URL for your local dev server Share your local dev server via a public HTTPS URL instantly. No VPN or staging needed—just two clicks to share with your team or clients. --- # One-click incident response operations deck — VibeControls Use Cases > Streamline incident response with VibeDeck. Execute common actions like API restarts and cache flushes with a single click. Source: https://vibecontrols.com/use-cases/vibedeck-incident-response ## One-click incident response operations deck Restart, check, rollback, alert — from a single VibeDeck Incident response at 3 AM is not the time to dig through a runbook for the restart command. Every second you spend reading docs is a second the service is down. VibeControls VibeDeck gives on-call engineers a pre-configured StreamDeck-style button grid for your operational actions. During normal hours, when you have context, you build the deck: add a COMMAND tile for "Restart API", a WEBHOOK tile for "Alert Slack #incidents", a SCRIPT tile that pulls the last 100 lines of the error log, another that rolls back the deploy by reverting the last container tag. Each tile runs on a specific agent — the right command on the right server. Output streams back to the browser in real time. If the restart succeeds, you see the green exit code. If not, you see the error immediately. Share the deck with your team via a public link with execution enabled. Now every on-call engineer has the same runbook-in-a-button, regardless of who's carrying the pager. Consistency reduces MTTR. ### Do it yourself Build a one-click incident response deck with command, script, and webhook tiles that run on a specific server agent, watch live output and exit codes, then share it with your team via a public execute-enabled link. - 1 Create your response actions /vibecontrols/actions Open Actions and create three actions: a COMMAND like "Restart API", a SCRIPT that pulls the last 100 lines of the error log, and a WEBHOOK that alerts your Slack channel. You should see: All three actions appear in your actions list, ready to be wired to deck tiles. [Open in app](https://app.vibecontrols.com/vibecontrols/actions) Mark done - 2 Create a VibeDeck /vibecontrols/vibedeck Go to VibeDeck and click create to spin up a new operations deck for your incident response. - 3 Add tiles bound to your actions /vibecontrols/vibedeck Use the add button to open the action picker, then pin a tile for each action: Restart API, pull error log, and alert Slack. - 4 Run a tile on a specific agent /vibecontrols/vibedeck Click a tile and pick the agent on the target server you want it to run on, then execute. - 5 Watch the output and exit code /vibecontrols/vibedeck Open the execution log panel to follow the run as it reports back to your browser. - 6 Share the deck with an execute link /vibecontrols/vibedeck Click "Public link", flip the toggle to Execute, set an expiry (4h, 24h, 7d, or never), and optionally add a password or a max-uses cap. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Centrally review scheduled database-backup runs across servers Self-serve operations for the whole org via shared VibeDeck ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # Platform team's shared ops deck eliminates 80% of tickets — VibeControls Use Cases > Reduce tickets by 80% with a shared VibeDeck. Enable product teams to self-serve operations while platform teams audit logs. Source: https://vibecontrols.com/use-cases/vibedeck-public-self-serve ## Platform team's shared ops deck eliminates 80% of tickets Safe, pre-approved operations — anyone can execute, everything is audited "Can you restart the staging API?" is a legitimate request. But when the platform team fields 20 variations of it per day, something has gone wrong in the access model. The right answer isn't to give everyone SSH keys — that creates a security nightmare. And it isn't to keep everything behind a ticket queue — that creates a velocity bottleneck. The right answer is governed self-service. VibeControls VibeDeck is the mechanism. Platform team builds a deck with the pre-approved, safe operations: restart by service name, flush cache by key pattern, query the error log for the last N lines, trigger a specific CI workflow. Each tile runs on a specific agent, with a specific command, with output captured. Enable "Public access" and "Allow execution" on the deck, paste the link in your internal wiki, and announce it in Slack. Product teams use it whenever they need an ops operation — no ticket, no DM to the platform team. The platform team's VibeDeck execution log shows who ran what, when, on which agent, with what output. The ticket queue shrinks. The audit trail grows. ### Do it yourself Build a deck of pre-approved per-agent ops tiles, share it as a public self-serve link that lets teammates run those operations safely, and review who ran what in the execution log. - 1 Create your ops deck /vibecontrols/vibedeck Open VibeDeck and create a new deck for your platform team's common operations. You should see: A fresh, empty deck appears, ready for you to add tiles. [Open in app](https://app.vibecontrols.com/vibecontrols/vibedeck) Mark done - 2 Add pre-approved operation tiles /vibecontrols/vibedeck Add a tile for each safe operation, choosing a Command, Script, or Webhook tile and pointing it at the specific agent and exact command you want it to run. - 3 Mint a public execute link /vibecontrols/vibedeck Click "Public link" on the deck, switch the toggle to Execute, set an expiry (4h, 24h, 7 days, or never), and optionally add a password or a maximum number of uses, then create the link. - 4 Share the link in your wiki /vibecontrols/vibedeck Copy the generated link and paste it into your internal wiki or team channel so product teams can find it. - 5 Run a tile from the public page Public Deck Link When a teammate opens the shared link, they see a clean public deck (commands and URLs stay hidden) and click a tile to run it. - 6 Review who ran what /vibecontrols/vibedeck Open the deck's execution log to see each run with its operation type, command, output, and exit or HTTP status. Ready to make this your story? ### Capabilities used ### Category For tech leads, engineering managers, and platform teams Golden-path templates that preserve service conventions Living software catalog with real ownership and dependency graphs ### Related use cases Team & Platform ### New hire productive in under an hour Onboard new hires in under an hour with VibeControls. Access shared sessions, environment variables, and documentation instantly. Team & Platform ### Golden-path templates that preserve service conventions Publish reusable component snapshots with variables and setup guidance, then instantiate consistent catalog components from them. Team & Platform ### Living software catalog with real ownership and dependency graphs Maintain a live software catalog with VibeControls. Register services, APIs, and libraries with ownership and dependency graphs.'s connected to the Vibes where work actually happens. --- # Send code to AI directly from VS Code — VibeControls Use Cases > Send code to AI directly from VS Code with VibeControls. Highlight, right-click, and get AI responses tailored to your active Vibe. Source: https://vibecontrols.com/use-cases/vscode-send-to-ai ## Send code to AI directly from VS Code Highlight, right-click, get an AI response grounded in your project context Legacy code is hard to understand precisely because you lack context: the architectural decisions, the constraints at the time, the business rules baked into the logic. AI can help, but only if it has the same context. The VibeControls VS Code extension solves this by connecting your editor to your active Vibe. When you right-click a highlighted function and choose "Send to AI", the extension pulls your project's AI Context (README, architecture docs, coding standards) and prepends it to the prompt before dispatching. The AI response is aware of your stack, your patterns, your constraints — not just the snippet you highlighted. It explains the function in terms of your actual architecture, not a generic description. You can also create and run sessions from VS Code without leaving your editor, start tunnels for whatever port your dev server is running on, and launch VibeDeck actions — all from the Explorer sidebar. The extension is a thin UI layer over the same VibeControls API, so nothing is duplicated: workspace, sessions, Vibes, and AI contexts are all shared with the web app and CLI. ### Do it yourself Highlight code in VS Code, send it to your AI provider with your active Vibe's project context attached, and read the grounded response right inside the editor. - 1 Install the VibeControls extension VS Code Open the Extensions view in VS Code, search for VibeControls, and click Install. You should see: The VibeControls extension is installed and its Explorer sidebar views appear. Mark done - 2 Sign in and pick your workspace VS Code Open the Command Palette and run VibeControls: Login, then choose your workspace when prompted. - 3 Open your project folder bound to a Vibe VS Code Open the folder for the project you want to work on. The extension automatically recognizes the active Vibe linked to that folder. - 4 Highlight the code you want help with VS Code Select a function or any block of code in the editor. - 5 Right-click and choose Send to AI VS Code Right-click the highlighted code and choose Send to AI from the context menu. - 6 Get answers grounded in your project context VS Code Nothing to do here — your active Vibe's AI Context is automatically attached and prepended to the prompt before it's sent. - 7 Read the response in the side panel VS Code View the themed response panel that opens in the editor. - 8 Run sessions, tunnels, and VibeDeck actions VS Code From the Command Palette, create or start a session, open a public tunnel for your dev-server port, or launch a VibeDeck action — all without leaving the editor. Ready to make this your story? ### Capabilities used ### Category For solo engineers and full-stack developers Shared AI prompt library with project context Give your AI assistant scoped access to your codebase ### Related use cases Individual Developer ### Zero-config local agent in under a minute Set up a VibeControls agent in under a minute. Transform your machine into a managed node with AI, sessions, and plugins—no config needed. Individual Developer ### Per-project environment variables without .env sprawl Manage environment variables per project with Vibe. Automatically apply secrets to sessions without .env files or risking production data. Individual Developer ### Any machine's terminal in a browser tab Open a live, interactive terminal to any of your machines — local or remote — straight from a browser tab. No SSH client, no key management friction, no VPN. --- # Instant Slack alert when a session fails — VibeControls Use Cases > Get instant Slack alerts for session failures. Subscribe to session.error events and notify your team without polling or dashboards. Source: https://vibecontrols.com/use-cases/webhook-slack-alert ## Instant Slack alert when a session fails Subscribe to session.error events and get a Slack message in seconds Long-running background sessions — data pipelines, AI training jobs, test runners — fail silently far too often. You find out an hour later when you check the output, not within seconds when there's still time to restart and recover. VibeControls webhooks change the notification model from pull to push. Create a webhook subscription: select session.error and session.stopped events, and point the endpoint at your Slack incoming webhook URL. VibeControls signs every payload with HMAC-SHA256 so your endpoint can verify authenticity. The payload includes the session name, agent hostname, error message, and exit code. Configure Slack to format and route it to the right channel — maybe #infra-alerts for production agents, #dev-channel for staging. From the moment a session fails, your team has the alert in Slack. The webhook delivery log in VibeControls shows every event — delivered, failed, retry status — so you can debug notification reliability too. Retry backoff (30s, 5m, 30m) means transient webhook endpoint failures don't lose events. ### Do it yourself Set up a signed webhook subscription so a Slack channel gets an instant alert whenever a VibeControls session fails or stops. - 1 Create a Slack incoming webhook Slack In Slack, add an Incoming Webhook app to the channel where you want alerts, then copy the generated webhook URL. You should see: You have a Slack webhook URL ready to paste into VibeControls. Mark done - 2 Open the Webhooks page /vibecontrols/webhooks From the VibeControls sidebar, click the Webhooks nav item to open the Webhooks page. - 3 Start a new webhook subscription /vibecontrols/webhooks Click New Webhook and paste your Slack webhook URL into the endpoint field. - 4 Subscribe to session failure events /vibecontrols/webhooks In the event picker, select the session error and session closed events so the webhook fires whenever a session fails or stops. - 5 Save the webhook /vibecontrols/webhooks Click Save to create the subscription. VibeControls generates a signing key and signs every payload with HMAC-SHA256. - 6 Send a test delivery /vibecontrols/webhooks Open your new webhook and click Test to send a sample payload to Slack. - 7 Watch live deliveries roll in /vibecontrols/webhooks Whenever a session errors or stops, return to the webhook's detail view to see its delivery activity, including delivery counts, the last delivery time, and any errors. Ready to make this your story? ### Capabilities used ### Category For platform, infrastructure, and reliability engineers Connect your database GUI to a remote database over a tunnel Queue AI test-generation jobs and review results in the morning ### Related use cases DevOps / SRE ### Bootstrap a multi-server agent fleet in under 5 minutes Quickly bootstrap a multi-server agent fleet with VibeControls. Automate agent installation and configuration across all SSH servers. DevOps / SRE ### Visual map of your entire infrastructure topology Visualize your infrastructure with the Agent Graph. See connectivity status and agent reachability for development and staging setups. DevOps / SRE ### Centrally review scheduled database-backup runs across servers Audit scheduled database backups across servers with Vibe. Centralized logging ensures you know every backup's status at a glance. --- # VibeControls vs Competitors | Consolidated Comparison > Compare VibeControls with every tracked competitor in one feature and pricing matrix, then open each detailed competitor page for deeper analysis. Source: https://vibecontrols.com/vs ## VibeControls vs the field Review every competitor together in one table, then drill into the dedicated /vs pages for the full analysis. ### Competitor-specific deep dives Each card links to the existing detailed comparison page. #### Warp The terminal is just the start. VibeControls is your workspace-native control plane. #### Termius A great SSH vault is not a control plane. #### Backstage A portal framework is not a runtime. #### Port The control plane that brings the catalog to your machines. #### GitHub Codespaces Flat-rate hybrid control. No cloud lock-in. #### Coder Hybrid control plane without the container-orchestration tax. ### What makes VibeControls different Every differentiator below is checked against what is actually built. Filter by delivery status, then open any card for the detail. ### Feature matrix Every tracked feature or capability from the competitor-specific pages, consolidated across all competitors. | Feature / capability | VibeControls | Warp | Termius | Backstage | Port | GitHub Codespaces | Coder | | --- | --- | --- | --- | --- | --- | --- | --- | | Core terminal / session | Browser-native, multi-provider | Native Rust terminal | Not listed | Not listed | Not listed | Not listed | Not listed | | Agent runtime on your targets | Self-installing agent | Oz cloud agents (Warp-hosted) | Not offered | Not offered | Not offered | Not listed | Not listed | | Agent fleet / topology | Visual Agent Graph | Single pane for local + Oz | Not offered | Not listed | Not listed | Not listed | Not listed | | AI orchestration | 15+ BYOK providers + MCP server | BYOK on Build+, multi-harness | Not listed | Not listed | Not listed | Not listed | Not listed | | Plan mode / HITL checkpoints | Plan mode + Plannotator | /plan mode | Not listed | Not listed | Not listed | Not listed | Not listed | | IDP layer (catalog, scorecards) | Available (initial) | Not offered | Not listed | Not listed | Not listed | Not listed | Not listed | | Governed / metered tunnels | VibeTunnels + Cloudflare | Not offered | Port forwarding only | Not listed | Not listed | Not listed | Not listed | | Native mobile app | iOS + Android | Web viewer only | iOS + Android (full SSH) | Not listed | Web only | Not offered | Not offered | | Pricing model | Flat-rate, regional tiers | Credit/usage-based | Not listed | Not listed | Not listed | Per-hour compute + storage | OSS free + Premium + infra | | Compliance certs | In flight | SOC 2 Type II | Not listed | Not listed | Not listed | SOC 2 / ISO / FedRAMP | Customer-controlled / Premium | | Core protocols | SSH + tunnel providers | Not listed | SSH/SFTP/Mosh/Telnet/Serial | Not listed | Not listed | Not listed | Not listed | | Real-time session sharing | RBAC-controlled share links | Not listed | Terminal Multiplayer (WebRTC) | Not listed | Not listed | Not listed | Not listed | | Multi-provider BYOK AI | 15+ providers + MCP | Not listed | Not offered | Not listed | Not listed | Not listed | Not listed | | IDP layer | Available (initial) | Not listed | Not offered | Not listed | Not listed | Not offered | Not offered | | Plugin/SDK ecosystem | Plugin SDK + Store (early catalog, growing) | Not listed | Snippets + integrations | Not listed | Not listed | Not listed | Not listed | | Team pricing | Team $29/user/mo annual | Not listed | Team $20/user/mo | Not listed | Basic $30/seat/mo (50-seat pkg) | Not listed | Not listed | | Software catalog | Available (initial) | Not listed | Not listed | Core strength | Flexible Context Lake | Not listed | Not listed | | Scorecards / standards | Available (Bronze/Silver/Gold) | Not listed | Not listed | Via TechInsights / Soundcheck | Scorecards + initiatives | Not listed | Not listed | | Golden-path templates | Vibe Deck + templates | Not listed | Not listed | Scaffolder | Not listed | Not listed | Not listed | | Terminal sessions | Multi-provider, browser-native | Not listed | Not listed | Not offered | Not offered | VS Code cloud terminal | Web IDE / SSH terminal | | Governed tunnels | VibeTunnels + Cloudflare | Not listed | Not listed | Not offered | Not offered | Port forwarding only | Port forwarding only | | BYOK AI orchestration | 15+ providers + MCP | Not listed | Not listed | Not offered | Not listed | Copilot / multi-model, no BYOK | BYO AI agent inside workspace | | Deployment model | Managed SaaS | Not listed | Not listed | Self-hosted OSS / managed options | Not listed | Cloud only (GitHub infra) | Self-hosted / any cloud / air-gapped | | Time to value | Minutes | Not listed | Not listed | Weeks to months | Not listed | Not listed | Not listed | | Pricing transparency | Flat per-seat tiers | Not listed | Not listed | OSS free / managed custom | Not listed | Not listed | Not listed | | Self-service actions | Vibe Deck + actions | Not listed | Not listed | Not listed | Actions + workflows | Not listed | Not listed | | BYOK AI for coding assistants | 15+ providers + MCP | Not listed | Not listed | Not listed | BYO-LLM for portal agents | Not listed | Not listed | | Free tier | 1 user, full surface | Not listed | Not listed | Not listed | 15 seats, 10K entities | Not listed | Not listed | | Target ownership | Your machines | Not listed | Not listed | Not listed | Not listed | GitHub/Microsoft hosted | Your cloud or container platform | | Agent runtime | Self-installing agent | Not listed | Not listed | Not listed | Not listed | Not offered | Workspace provisioners | ### Pricing and operating model matrix Pricing, packaging, implementation, and commercial aspects shown side by side where the source data includes them. | Pricing / aspect | VibeControls | Warp | Termius | Backstage | Port | GitHub Codespaces | Coder | | --- | --- | --- | --- | --- | --- | --- | --- | | Free tier | $0 — full surface, 1 agent | $0 — limited Oz + Warp Drive | $0 — core SSH local vault | Not listed | $0 — up to 15 seats | $0 — 120 core-hrs + 15 GB/mo | $0 — Community Edition (self-hosted infra) | | Individual paid | Hobby $9/mo | Build $20/mo | Pro $10/mo | Not listed | N/A | Not listed | N/A (OSS) | | Pro / power user | Pro $29/mo flat | Build/Max $20–$200/mo (credits) | N/A (Pro is individual) | Not listed | N/A | Not listed | N/A (OSS) | | Team entry | Team $29/user/mo annual | Business $50/user/mo | Team $20/user/mo (annual) | Not listed | Basic $30/seat/mo (50 seats) | Not listed | Premium ~$40/user/mo+ (plus infra) | | Regional pricing | Four-band (USD/INR/etc.) | USD only | Not listed | Not listed | Not listed | USD only | USD only | | Runtime included | Agents, sessions, tunnels, AI | Not listed | SSH vault only | Requires custom plugins | Portal only | Not listed | Not listed | | License | SaaS subscription | Not listed | Not listed | OSS $0 | Not listed | Not listed | Not listed | | 10 devs, first year | ~$3,480 (Team) | Not listed | Not listed | ~$78K–160K (OSS + FTE) | Not listed | Not listed | Not listed | | Managed option | Included SaaS | Not listed | Not listed | Roadie ~$18K–30K/yr + seats | Not listed | Not listed | Not listed | | Maintenance | Zero (SaaS) | Not listed | Not listed | Significant (OSS) | Not listed | Not listed | Not listed | | Pro flat rate | Pro $29/mo | Not listed | Not listed | Not listed | Not listed | ~$29–$115/mo full-time (2–8 core) | Not listed | | 10 devs, full-time 4-core, annual | ~$3,480 (Team) | Not listed | Not listed | Not listed | Not listed | ~$6,912+ before storage | Not listed | | Self-hosting | Enterprise add-on | Not listed | Not listed | Not listed | Not listed | Not available | Core capability (OSS + Premium) | ### Need a competitor added? Tell us what you are evaluating and we will expand the comparison set. --- # VibeControls vs Backstage | VibeControls plugin > Compare VibeControls and Backstage across catalog, scorecards, templates, runtime, agents, sessions, and total cost of ownership. Source: https://vibecontrols.com/vs/backstage ## VibeControls vs Backstage A portal framework is not a runtime. Backstage is the leading open-source IDP framework. VibeControls gives you the developer control plane — terminal sessions, agents, AI — that Backstage cannot reach without heavy custom work. ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | Backstage | | --- | --- | --- | | Software catalog | Available (initial) | Core strength | | Scorecards / standards | Available (Bronze/Silver/Gold) | Via TechInsights / Soundcheck | | Golden-path templates | Vibe Deck + templates | Scaffolder | | Agent runtime on your targets | Self-installing agent | Not offered | | Terminal sessions | Multi-provider, browser-native | Not offered | | Governed tunnels | VibeTunnels + Cloudflare | Not offered | | BYOK AI orchestration | 15+ providers + MCP | Not offered | | Deployment model | Managed SaaS | Self-hosted OSS / managed options | | Time to value | Minutes | Weeks to months | | Pricing transparency | Flat per-seat tiers | OSS free / managed custom | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | Backstage | | --- | --- | --- | | License | SaaS subscription | OSS $0 | | 10 devs, first year | ~$3,480 (Team) | ~$78K–160K (OSS + FTE) | | Managed option | Included SaaS | Roadie ~$18K–30K/yr + seats | | Runtime included | Agents, sessions, tunnels | Requires custom plugins | | Maintenance | Zero (SaaS) | Significant (OSS) | ### Why VibeControls wins Honest differentiators — and where Backstage still has an edge. #### Execution layer included Backstage is a portal framework with no agents, sessions, tunnels, or sandboxes. VibeControls ships a live runtime on your targets — a self-installing agent fleet with a visual Agent Graph, browser-native terminal sessions, and governed tunnels. #### SaaS time-to-value Backstage OSS requires weeks to months of setup. VibeControls is SaaS: workspace, agent, session in minutes. #### Predictable TCO Backstage is free to license but expensive to operate. VibeControls replaces hidden engineering cost with a flat per-seat fee. #### Governed BYOK AI with human-in-the-loop A BYOK AI hub (Claude, Codex, Gemini, OpenRouter, Ollama, and any OpenAI-compatible endpoint) plus an MCP server, Plan mode, and the Plannotator review portal for reviewable checkpoints before execution. Backstage has no native AI execution layer. ### Honest concessions Backstage is a strong product. Here is where it leads today. - Backstage's Software Catalog is more mature than VibeControls' initial catalog. - Backstage TechDocs is a proven docs-as-code solution. - Backstage Templates (Scaffolder) are deeper for golden-path service creation today. - Backstage OSS is free to license if you have engineering capacity. ### Ready to switch from Backstage? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # VibeControls vs Coder | VibeControls plugin > Compare VibeControls and Coder across self-hosted vs hybrid, local targets, mobile, tunnels, AI, and pricing. Source: https://vibecontrols.com/vs/coder ## VibeControls vs Coder Hybrid control plane without the container-orchestration tax. Coder is the leading open-source, self-hosted cloud development environment. VibeControls gives you a hybrid control plane across local, remote, and cloud targets — with no container orchestration required. ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | Coder | | --- | --- | --- | | Deployment model | Local + remote + cloud, managed SaaS | Self-hosted / any cloud / air-gapped | | Target ownership | Your machines | Your cloud or container platform | | Agent runtime | Self-installing agent | Workspace provisioners | | Terminal sessions | Multi-provider, browser-native | Web IDE / SSH terminal | | Governed tunnels | VibeTunnels + Cloudflare | Port forwarding only | | BYOK AI orchestration | 15+ providers + MCP | BYO AI agent inside workspace | | IDP layer | Available (initial) | Not offered | | Native mobile app | iOS + Android | Not offered | | Pricing model | Flat per-seat SaaS | OSS free + Premium + infra | | Compliance certs | In flight | Customer-controlled / Premium | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | Coder | | --- | --- | --- | | Free tier | $0 — 1 agent, full surface | $0 — Community Edition (self-hosted infra) | | Individual paid | Hobby $9/mo | N/A (OSS) | | Pro / power user | Pro $29/mo | N/A (OSS) | | Team entry | Team $29/user/mo annual (5+ seats) | Premium ~$40/user/mo+ (plus infra) | | Self-hosting | Enterprise add-on | Core capability (OSS + Premium) | | Regional pricing | Four-band (USD/INR/etc.) | USD only | ### Why VibeControls wins Honest differentiators — and where Coder still has an edge. #### Hybrid freedom Coder is cloud-workspace centric. VibeControls runs on your laptop, remote VM, or cloud host — your choice. #### No container orchestration required Coder commonly relies on a container-orchestration layer. VibeControls uses lightweight agents and managed SaaS, removing cluster ops. #### Terminal-first control plane VibeControls treats browser-native terminal sessions as a first-class primitive, not just an IDE convenience. #### Native mobile apps Approve tunnels, check agents, and manage sessions from iOS and Android. Coder is desktop/web focused. ### Honest concessions Coder is a strong product. Here is where it leads today. - Coder gives customers full control of infrastructure and data residency by default. - Coder's open-source AGPL core is free to audit and extend. - Coder has deeper IDE integrations (JetBrains, Cursor) out of the box. - Coder is purpose-built for large-scale standardized cloud workspaces. ### Ready to switch from Coder? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # VibeControls vs GitHub Codespaces | VibeControls plugin > Compare VibeControls and GitHub Codespaces across cloud vs hybrid, pricing, AI, tunnels, compliance, and developer control. Source: https://vibecontrols.com/vs/codespaces ## VibeControls vs GitHub Codespaces Flat-rate hybrid control. No cloud lock-in. GitHub Codespaces is a GitHub-native cloud IDE. VibeControls works on local, remote, and cloud targets, with flat-rate pricing that does not punish full-time use. ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | GitHub Codespaces | | --- | --- | --- | | Deployment model | Local + remote + cloud | Cloud only (GitHub infra) | | Target ownership | Your machines | GitHub/Microsoft hosted | | Agent runtime | Self-installing agent | Not offered | | Terminal sessions | Multi-provider, browser-native | VS Code cloud terminal | | Governed tunnels | VibeTunnels + Cloudflare | Port forwarding only | | BYOK AI orchestration | 15+ providers + MCP | Copilot / multi-model, no BYOK | | IDP layer | Available (initial) | Not offered | | Native mobile app | iOS + Android | Not offered | | Pricing model | Flat per-seat | Per-hour compute + storage | | Compliance certs | In flight | SOC 2 / ISO / FedRAMP | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | GitHub Codespaces | | --- | --- | --- | | Free tier | $0 — 1 agent, full surface | $0 — 120 core-hrs + 15 GB/mo | | Pro flat rate | Pro $29/mo | ~$29–$115/mo full-time (2–8 core) | | 10 devs, full-time 4-core, annual | ~$3,480 (Team) | ~$6,912+ before storage | | Self-hosting | Enterprise add-on | Not available | | Regional pricing | Four-band (USD/INR/etc.) | USD only | ### Why VibeControls wins Honest differentiators — and where GitHub Codespaces still has an edge. #### Hybrid freedom Codespaces is cloud-only. VibeControls runs on your laptop, remote VM, or cloud host — your choice. #### Flat-rate for full-time teams A full-time 4-core codespace costs ~$57.60/mo compute-only. VibeControls Pro is a flat $29/mo. #### Workspace-native control plane Codespaces has no catalog, scorecards, or agent fleet graph. VibeControls wraps runtime and an initial IDP together. #### Multi-provider BYOK AI Bring your own keys for Claude, Codex, Gemini, and more. Codespaces ties AI to Copilot. ### Honest concessions GitHub Codespaces is a strong product. Here is where it leads today. - Codespaces' GitHub integration is tighter because it is a GitHub product. - Codespaces has a larger extension ecosystem through VS Code. - Codespaces can be cheaper for occasional usage and has a strong free tier. - Codespaces has formal compliance certifications today. ### Ready to switch from GitHub Codespaces? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # VibeControls vs Port — Internal Developer Portal Alternative > Compare VibeControls and Port across software catalog, scorecards, self-service actions, agents, sessions, tunnels, AI, and pricing. Source: https://vibecontrols.com/vs/port ## VibeControls vs Port The control plane that brings the catalog to your machines. Port is a strong SaaS IDP. VibeControls starts where developers actually work — on their machines — and brings catalog, governance, and AI along for the ride. ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | Port | | --- | --- | --- | | Software catalog | Available (initial) | Flexible Context Lake | | Scorecards / standards | Available (Bronze/Silver/Gold) | Scorecards + initiatives | | Self-service actions | Vibe Deck + actions | Actions + workflows | | Agent runtime on your targets | Self-installing agent | Not offered | | Terminal sessions | Multi-provider, browser-native | Not offered | | Governed tunnels | VibeTunnels + Cloudflare | Not offered | | BYOK AI for coding assistants | 15+ providers + MCP | BYO-LLM for portal agents | | Native mobile app | iOS + Android | Web only | | Free tier | 1 user, full surface | 15 seats, 10K entities | | Team pricing | Team $29/user/mo annual | Basic $30/seat/mo (50-seat pkg) | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | Port | | --- | --- | --- | | Free tier | $0 — 1 agent, full surface | $0 — up to 15 seats | | Individual paid | Hobby $9/mo | N/A | | Pro / power user | Pro $29/mo | N/A | | Team entry | Team $29/user/mo annual (5+ seats) | Basic $30/seat/mo (50 seats) | | Runtime included | Agents, sessions, tunnels, AI | Portal only | ### Why VibeControls wins Honest differentiators — and where Port still has an edge. #### Starts where developers work Port is a great catalog, but it stops at the portal. VibeControls installs agents on your machines and brings governance along. #### Agent runtime + sessions + tunnels VibeControls provides agent fleet management, browser-native terminals, governed tunnels, and sandboxes. Port has no runtime layer. #### Lower per-seat price with more surface VibeControls Team is $29/user/mo annual. Port Basic starts at $30/seat/mo for a 50-seat package and is portal-only. #### Multi-provider BYOK AI hub VibeControls orchestrates coding assistants with workspace audit. Port AI is portal-workflow focused. ### Honest concessions Port is a strong product. Here is where it leads today. - Port's software catalog is more mature than VibeControls' initial catalog. - Port's scorecards and initiatives are purpose-built and real-time. - Port's self-service actions framework is deeper for portal-centric workflows. - Port's 15-seat free tier is generous for team evaluation. ### Ready to switch from Port? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # VibeControls vs Termius — SSH Client Alternative > Compare VibeControls and Termius across SSH sessions, team vaults, real-time collaboration, agents, tunnels, AI, and pricing. Source: https://vibecontrols.com/vs/termius ## VibeControls vs Termius A great SSH vault is not a control plane. Termius is a polished cross-platform SSH client and encrypted vault. VibeControls adds a workspace-native control plane around those sessions: agents, AI orchestration, governed tunnels, and an IDP (initial but available). ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | Termius | | --- | --- | --- | | Core protocols | SSH + tunnel providers | SSH/SFTP/Mosh/Telnet/Serial | | Agent runtime on your targets | Self-installing agent | Not offered | | Agent fleet / topology | Visual Agent Graph | Not offered | | Real-time session sharing | RBAC-controlled share links | Terminal Multiplayer (WebRTC) | | Multi-provider BYOK AI | 15+ providers + MCP | Not offered | | IDP layer | Available (initial) | Not offered | | Governed / metered tunnels | VibeTunnels + Cloudflare | Port forwarding only | | Native mobile app | iOS + Android | iOS + Android (full SSH) | | Plugin/SDK ecosystem | Plugin SDK + Store (early catalog, growing) | Snippets + integrations | | Team pricing | Team $29/user/mo annual | Team $20/user/mo | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | Termius | | --- | --- | --- | | Free tier | $0 — 1 agent, full surface | $0 — core SSH local vault | | Individual paid | Hobby $9/mo | Pro $10/mo | | Pro / power user | Pro $29/mo | N/A (Pro is individual) | | Team entry | Team $29/user/mo annual (5+ seats) | Team $20/user/mo (annual) | | Runtime included | Agents, sessions, tunnels, AI | SSH vault only | ### Why VibeControls wins Honest differentiators — and where Termius still has an edge. #### Agent-based control plane with fleet topology Termius stores credentials. VibeControls installs an agent that turns every machine into a managed, auditable node — rendered live in a visual Agent Graph of targets, reachability, and active sessions. Agents can even self-register from a tunnel URL. #### Workspace-native IDP layer Termius has no catalog, scorecards, or golden paths. VibeControls ships an initial IDP with catalogs, scorecards, and Vibe Deck templates. #### Multi-provider BYOK AI VibeControls ships a BYOK AI hub with 15+ providers and an MCP server. Termius has no AI orchestration layer. #### More surface per dollar VibeControls Team bundles agents, sessions, tunnels, AI, and governance. Termius Team covers SSH vault sharing only. ### Honest concessions Termius is a strong product. Here is where it leads today. - Termius native mobile/tablet SSH UX is excellent for pure server administration. - Termius Multiplayer is quick to start via peer-to-peer WebRTC. - Termius has a lower entry price for individuals who only need cross-device SSH sync. - Termius supports protocols VibeControls does not natively (Telnet, Serial, Mosh). ### Ready to switch from Termius? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # VibeControls vs Warp — Agentic Engineering Mission Control > Compare VibeControls vs Warp for terminal sessions, AI orchestration, governed tunnels, IDP, and mobile. Discover workspace-native control benefits. Source: https://vibecontrols.com/vs/warp ## VibeControls vs Warp The terminal is just the start. VibeControls is your workspace-native control plane. Warp is a powerful, AI-native terminal. VibeControls is a developer control plane that wraps agents, sessions, tunnels, AI orchestration, and an IDP (catalog, scorecards, templates — initial but available) into one workspace — with execution staying on your targets. ### Feature comparison Side-by-side capabilities that matter for engineering teams. | Capability | VibeControls | Warp | | --- | --- | --- | | Core terminal / session | Browser-native, multi-provider | Native Rust terminal | | Agent runtime on your targets | Self-installing agent | Oz cloud agents (Warp-hosted) | | Agent fleet / topology | Visual Agent Graph | Single pane for local + Oz | | AI orchestration | 15+ BYOK providers + MCP server | BYOK on Build+, multi-harness | | Plan mode / HITL checkpoints | Plan mode + Plannotator | /plan mode | | IDP layer (catalog, scorecards) | Available (initial) | Not offered | | Governed / metered tunnels | VibeTunnels + Cloudflare | Not offered | | Native mobile app | iOS + Android | Web viewer only | | Pricing model | Flat-rate, regional tiers | Credit/usage-based | | Compliance certs | In flight | SOC 2 Type II | ### Pricing comparison Transparent, predictable pricing for teams that ship every day. | Plan / cost area | VibeControls | Warp | | --- | --- | --- | | Free tier | $0 — full surface, 1 agent | $0 — limited Oz + Warp Drive | | Individual paid | Hobby $9/mo | Build $20/mo | | Pro / power user | Pro $29/mo flat | Build/Max $20–$200/mo (credits) | | Team entry | Team $29/user/mo annual | Business $50/user/mo | | Regional pricing | Four-band (USD/INR/etc.) | USD only | ### Why VibeControls wins Honest differentiators — and where Warp still has an edge. #### Workspace-native IDP, not just a smarter terminal Warp has no software catalog, scorecards, or golden-path scaffolding. VibeControls ships an initial IDP (catalogs, Bronze/Silver/Gold scorecards, Vibe Deck templates) so the same workspace runs and governs your engineering operation. #### Execution stays on your targets VibeControls agents run on your laptop, VM, cloud host, or container. Warp Oz runs in Warp-hosted infrastructure by default. #### Governed, metered tunnels Tunnels are first-class workspace resources: reserve a managed *.vibetunnels.com subdomain (or bring a custom domain), gate access by RBAC, and meter starts and active hours for billing. Warp has no equivalent developer-tunnel primitive. #### Flat, regional, predictable pricing VibeControls Pro is a flat $29/mo. Warp Build is $20/mo (1,500 credits) and Max reaches $200/mo — credit-based, so heavy cloud-agent usage can spike the bill. ### Honest concessions Warp is a strong product. Here is where it leads today. - Warp's terminal UX is stronger for day-to-day command-line work. - Warp's live multiplayer UX (avatars, cursors, live agent traces) is richer today. - Warp already holds SOC 2 Type II. - Warp is more widely adopted as a standalone terminal brand. ### Ready to switch from Warp? Start free on VibeControls and see why teams choose the workspace-native control plane. --- # Join VibeControls Waitlist > Be the first to access VibeControls. Sign up now for early access and stay ahead with our innovative AI solutions. Source: https://vibecontrols.com/waitlist ## VibeControls is now available! Great news — VibeControls is open to everyone. Sign up or log in to get started.