Agent Fleet and Agent Detail
Browse every agent in the workspace as filterable cards, then open one to inspect its connectivity, plugins, sessions and logs.
These images are illustrations of the concept, not screenshots of the actual product.
Overview
An agent is the runtime installed on a machine you develop or operate on, and in VibeControls it is the unit everything else hangs from. This concept covers the two views a person moves between constantly: the list of every agent in the workspace, and the page for a single agent where its real state lives.
The problem is visibility at two scales. Across a fleet you want to know quickly which agents answered recently and which did not, without reading a wall of text. Inside one agent you want considerably more, because an agent that reports as connected can still be running a degraded plugin, holding stale sessions or filling its log with errors. The illustrated design separates those scales rather than compromising with one crowded table.
The landing illustration shows the agents view as a grid of cards, with a count of how many agents are online, a refresh control and an action to add another. A short banner explains what agents actually do. A search field, a grid and list toggle, and filters for status, platform and tag sit above the cards. Each card carries the agent name, its host, a status pill, how long ago it was last seen, the machine architecture, a truncated connection URL and an overflow menu, so the fleet reads as a set of comparable tiles.
The agent detail illustration opens one of those cards. The header repeats the agent name with an enabled badge, a version badge and the host name, and offers a reconnect action. A wide tab row divides the agent into overview, profiles, security, sessions, plugins, logs, connection log, projects, notes, backup and maintenance, with counts on the tabs that hold items. The plugins tab is shown, split into installed, available and interface plugins, with a search field and a refresh control above a list of plugin rows.
What the plugin list demonstrates is the governance idea behind the product: each row carries a name, a version and one or two state pills, so an update that is waiting, a plugin that is degraded, one that is disconnected and one that is erroring are all legible at a glance. Selection checkboxes on each row imply acting on several plugins at once, which connects this surface to the wider fleet rollout and automation parts of VibeControls.
What this concept shows
- An agents view with a count of how many agents are currently online and a refresh control
- Card and list display modes with a search field and filters for status, platform and tag
- Agent cards showing name, host, status pill, last-seen time, machine architecture and a truncated connection URL
- A per-agent header with enabled state, version badge, host name and a reconnect action
- A wide tab row covering overview, profiles, security, sessions, plugins, logs, connection log, projects, notes, backup and maintenance, with counts on the populated tabs
- A plugins tab split into installed, available and interface plugins, each with its own count
- Plugin rows carrying a version and paired state pills for enabled, degraded, disconnected, idle, error and update-available conditions
- Row-level checkboxes that imply acting on several plugins in one step
How it works
- Open the agents view and read how many agents are online against the total.
- Narrow the grid with search or the status, platform and tag filters to find the machine you care about.
- Compare cards on last-seen time, architecture and status to spot an agent that has drifted.
- Open an agent to see its enabled state, version and host, and reconnect it if needed.
- Move to the plugins tab and review installed plugins for version drift, degraded state or errors.
- Select one or more plugin rows to act on them, or switch to the sessions, logs or connection log tabs to investigate further.
Who it's for
- Developers managing several machines from one workspace
- DevOps and platform engineers operating a fleet of agents
- Site reliability engineers triaging a degraded or unreachable machine
- Team leads auditing which agent tooling is installed where
Illustrations
2 illustrations of this concept. Select one to view it full size.
Agent Fleet Grid
This illustration shows the agents view with three tabs across the top for targets, agents and a graph view, with agents selected. The heading pairs the section title with a count of how many agents are online against the total, beside a refresh control and an action to add an agent. A short banner explains that agents run terminal sessions on machines you own. Below it sit a search field, a toggle between grid and list layouts, and dropdown filters for status, platform and tag. Six agent cards fill the grid, each with an icon, the agent name, the host beneath it, a status pill reading active, degraded or offline, a last-seen time, the machine architecture, a truncated connection URL and an overflow menu. The machine names, architectures and URLs shown are sample data.
Agent Detail and Plugins
This illustration shows the page for a single agent. A breadcrumb traces from the agents section to this machine, and the header pairs the agent name with an enabled badge, a version badge and the host name, alongside a reconnect action and an overflow menu. A wide tab row divides the agent into overview, profiles, security, sessions, plugins, logs, connection log, projects, notes, backup and maintenance, with numeric counts on the sessions and plugins tabs. The plugins tab is open, filtered into installed, available and interface groups, each carrying its own count, beside a plugin search field and a refresh control. Six plugin rows follow, each with a selection checkbox, a monospace plugin name, a version and one or two state pills covering enabled, update available, degraded, active, offline, disconnected, idle and error conditions. The plugin names and versions are sample data.
Topics
- agent fleet management
- agent status monitoring
- developer agent detail page
- installed plugin versions
- agent connectivity troubleshooting
- filter agents by platform
- agent last seen time
- plugin update available
- agent session and log tabs
- multi machine developer tooling
Related concepts

Targets and Agent Mesh Topology
Register the machines your agents can reach, inspect one in detail, and see the whole agent-to-agent mesh as a graph.
3 illustrations
Plugin Marketplace and Fleet Rollout
Browse plugins that extend every agent, check prerequisites and versions, then roll one out across chosen machines in a single step.
4 illustrations
Install and Connect Your First Agent
A guided path from a one-line install command on your own machine to a registered, connected agent in your workspace.
2 illustrations
Persistent Browser Terminal Sessions
Keep long-running terminals alive on your own machines, then find, watch and control them from any browser.
2 illustrations