$ loading_
Launch, monitor, and control Claude Managed Agents from any MCP client.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "managed-agent-control-mcp" yet — see the docs or source repo.
Use managed-agent-control-mcp to launch a Claude Managed Agent to analyze why this repository's CI is failing, and continuously return its event stream and current status.
A successfully launched agent session plus subsequent events, status updates, and interim analysis findings.
Watch this agent's tool requests; when it asks to run shell commands and read logs, show me the approval details first, continue only after approval, and send the results back to the agent.
A record of pending approvals, post-approval execution results, and the agent continuing the task based on those results.
If the agent gets stuck or needs more context, send follow-up guidance on my behalf; if the task drifts off target, stop the agent immediately and return a full event summary up to termination.
A successful human reply sent to the agent, or a terminated session with final status and a complete event summary.
Query Claude Code transcript analytics for cost, safety, audit, and efficiency insights.
Manage MCP servers in the Claude app through a simple web UI.
Let AI control macOS apps, mouse, and keyboard to automate tasks.
Manage MCP servers in Claude App and discover or install packages.
Wrap Claude Code CLI as MCP tools for headless coding sessions and automation.
Run a production-ready MCP server for files, HTTP, system info, and environment tools.