plane-agent¶
Plane project-management MCP server + A2A agent for the agent-utilities ecosystem — typed, deterministic tools over the Plane REST API for projects, work items, cycles, modules, and the rest of the Plane work-management surface.
Official documentation
This site is the canonical reference for plane-agent, maintained alongside every
release.
Overview¶
plane-agent wraps the Plane REST API with typed,
deterministic MCP tools and ships an optional A2A agent server that drives those
tools conversationally. It provides:
Api— a composed REST client (plane_agent.api_client.Api) over the Plane workspace API, organized by domain (projects, work items, cycles, modules, states, initiatives, intake, milestones, workspaces).- A broad MCP tool surface — action-routed tool domains for every Plane
resource, gated individually by
*TOOLenvironment toggles. - An A2A agent server (
plane-agentconsole script) that connects to the MCP server overMCP_URLand exposes the tools to a Pydantic-AI agent.
The agent remains inactive when credentials are absent — a PLANE_API_KEY
and PLANE_WORKSPACE_SLUG are required before any tool will execute.
Explore the documentation¶
- Installation — pip, source, uv, and the prebuilt Docker image.
- Deployment — run the MCP server and agent, Docker Compose, Caddy + Technitium.
- Usage — the MCP tools, the
Apiclient, and the CLI. - Backing Platform — deploy a self-hosted Plane instance with Docker.
- Overview — ecosystem role, enterprise posture, and architecture.
- Concepts — the
CONCEPT:PLANE-*registry.
Quick start¶
Connect it to a Plane workspace:
export PLANE_BASE_URL=https://api.plane.so
export PLANE_API_KEY=your_plane_api_key
export PLANE_WORKSPACE_SLUG=your-workspace
plane-mcp --transport streamable-http --host 0.0.0.0 --port 8000
See Installation and Deployment for the full matrix (PyPI install, Docker image, all transports, the agent server, reverse proxy, DNS).