Agent Fleet and Agent Detail
See every agent's status, platform and tunnel at a glance, then drill into one agent to manage its plugins, sessions and logs.
These images are illustrations of the concept, not screenshots of the actual product.
Overview
Once agents are installed, the next need is to manage them as a fleet. This concept shows two connected views in VibeControls: the Agents tab of Targets & Agents, which lists every agent in the workspace, and the agent detail page, which gathers everything about a single agent in one place. Together they are designed to replace ad hoc SSH checks with a shared, current picture of which agents exist, whether they are healthy and what they are running.
The agents list illustration presents the fleet as a grid of cards. Each card carries the agent's name and host, the operating system, a status pill such as Active, Degraded or Offline, how recently it was seen, its CPU architecture and the tunnel address it is reachable through. A header summary counts how many agents are online, and search, a grid or list toggle and filters for status, platform and tag make it practical to narrow a large fleet. Refresh and Add Agent actions sit at the top, next to tabs for Targets and the Agent Graph.
The agent detail illustration focuses on one machine. Its header shows the agent's enabled state, version and local hostname, with a Reconnect action for recovering a dropped link. A row of tabs covers overview, profiles, security, sessions, plugins, logs, connection history, bound vibes, notes, backup and maintenance. The Plugins tab is open, separating installed plugins from those available to add and those that contribute interface panels. Each installed plugin, from coding agents to terminal and session providers, shows its version and paired status badges, so a pending update, a degraded integration or an error stands out immediately.
This view is where fleet-level problems become fixable. An offline card on the list leads to the agent page, where connection logs, plugin state and maintenance actions sit side by side, and plugin changes made here feed into the wider plugin marketplace and rollout flows.
What this concept shows
- Card grid of agents with status pill, operating system, last-seen time, CPU architecture and tunnel address
- Online count summary with refresh and Add Agent actions
- Search plus status, platform and tag filters, with grid and list layouts
- Agent detail header with enabled state, version badge, hostname and a Reconnect action
- Tabs for profiles, security, sessions, plugins, logs, connection log, vibes, notes, backup and maintenance
- Plugin list split into installed, available and interface plugins, each with counts
- Per-plugin version and paired status badges that flag updates, degradation, disconnection and errors
How it works
- Open Targets & Agents and switch to the Agents tab to see the fleet as cards with an online count.
- Filter by status, platform or tag, or search by name, to find the agents that need attention.
- Open an agent's card to reach its detail page and check its enabled state, version and hostname.
- Switch to the Plugins tab to review installed plugins, their versions and status badges.
- Act on what you find: reconnect the agent, update a plugin, or open its sessions, logs or connection log for deeper diagnosis.
Who it's for
- Developers who run coding agents on their own laptops and servers
- Platform and DevOps engineers maintaining a fleet of agents
- Team leads checking which agents and plugins their team relies on
- Support engineers diagnosing a disconnected or degraded agent
Illustrations
2 illustrations of this concept. Select one to view it full size.
Agents Tab with Fleet Cards
This illustration shows the Targets & Agents area with the Agents tab selected between Targets and Agent Graph. The heading reports how many agents are online out of the total, with a refresh button and an Add Agent button beside it. A banner explains that agents run terminal sessions on your own machines. A toolbar offers agent search, a toggle between grid and list layouts, and dropdown filters for status, platform and tag. The main area is a grid of six cards for sample machines such as a developer laptop, a production web server, a staging API, a GPU trainer, a CI runner and a notification service. Each card shows the agent name, its host with an operating system icon, a status pill reading Active, Degraded or Offline, a menu button, the time since it was last seen, the CPU architecture and a highlighted tunnel address.
Agent Detail with Installed Plugins
This illustration shows the detail page for a single agent, reached through a breadcrumb from Targets & Agents and Agents. The header shows a terminal icon, the agent name, an Enabled pill, a version badge and the agent's local hostname, with a Reconnect button and a menu on the right. A tab row lists Overview, Profiles, Security, Sessions with a count, Plugins with a count, Logs, Connection Log, Vibes, Notes, Backup and Maintenance. The Plugins tab is active and offers segments for installed, available and interface plugins, each with a count, plus a plugin search box and a refresh button. Six installed plugins are listed with checkboxes, including coding agents and terminal and session providers, each with a version number. Paired badges show states such as Enabled with Update available, Enabled with Degraded, Active with Offline, Disconnected with Idle, and Enabled with Error.
Topics
- agent fleet management
- coding agent status
- agent detail page
- agent plugin management
- reconnect remote agent
- agent health monitoring
- developer machine inventory
- tunnel status agents
- manage AI coding agents
Related concepts

Install and Connect Your First Agent
Install the agent with one command, start it, and connect it to your workspace in a short guided two-step flow.
2 illustrations
Targets and Agent Mesh Topology
Register the machines your agents can reach over SSH or directly, and see the agent-to-agent mesh as an interactive 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
Fleet Health and Live Logs
See which agents, sessions, vibes and targets are healthy, then stream a struggling agent's logs to find out why.
2 illustrations