Overview

This Plane MCP Server implements the Model Context Protocol (MCP) to expose Plane project and issue data to tools and LLM-driven agents. It acts as a thin adapter between the Plane REST API and any MCP-compatible client, providing structured tool calls for listing projects, querying issues, creating and updating issues, and other common project-management operations.

Running a local or hosted MCP server for Plane makes it straightforward to embed project/issue context into AI workflows, automate triage and updates, and let LLMs interact with your issue tracker in a controlled, auditable way. The server is self-hostable, configurable via environment variables, and designed to be plugged into an LLM orchestration or agent framework that understands MCP-style tool invocation.

Features

MCP-compliant server that wraps Plane’s API for use by tools and agents
Read and write access to Plane projects, issues, comments, and labels
Authentication via API key (configurable)
Self-hostable with simple environment-driven configuration
REST endpoints suitable for integration with LLM toolchains and automation scripts
Basic request/response shapes compatible with MCP-style tool invocation

Installation / Configuration

Prerequisites: Node.js (14+) or Docker. Clone the repository and configure environment variables before starting.

Clone and install

git clone https://github.com/kelvin6365/plane-mcp-server.git
cd plane-mcp-server
npm install

Development run

# set environment variables in your shell or a .env file
export PLANE_API_KEY="your_plane_api_key"
export PORT=3000

npm start
# or for TypeScript projects:
# npm run dev

Sample .env

PLANE_API_KEY=sk_abcdef123456
PLANE_BASE_URL=https://api.plane.inc      # optional, if using a custom host
PORT=3000
LOG_LEVEL=info

Docker

# build
docker build -t plane-mcp-server .

# run
docker run -e PLANE_API_KEY=sk_abcdef123456 -p 3000:3000 plane-mcp-server

Configuration options (typical)

PLANE_API_KEY — API key used to authenticate to Plane
PLANE_BASE_URL — Base URL for the Plane API (useful for self-hosted Plane instances)
PORT — Port the MCP server listens on
LOG_LEVEL — Logging verbosity

Available Tools / Resources

The server exposes MCP-style tools/resources that map to Plane concepts. Typical resources include:

MCP Tool / Resource	Plane Action
projects.list	List all projects visible to the API key
projects.get	Get project details by ID
issues.list	List issues in a project or across projects
issues.get	Retrieve a single issue with metadata and comments
issues.create	Create a new issue (title, description, labels, assignee)
issues.update	Update fields on an issue (status, assignee, labels)
comments.create	Add a comment to an issue
labels.list	List labels available for a project

Each tool accepts a JSON payload and returns structured JSON results suitable for downstream processing by LLMs or automation scripts.

Example: create an issue (curl)

curl -X POST http://localhost:3000/tools/issues.create \
  -H "Content-Type: application/json" \
  -d '{
    "projectId": "proj_123",
    "title": "Bug: API returns 500 for X",
    "description": "Steps to reproduce: ...",
    "labels": ["bug","high-priority"],
    "assignee": "user_456"
  }'

Use Cases

Automating triage: have an LLM classify incoming issues, add labels and assign priority via issues.update and labels.create calls.
Synthesizing status reports: periodically call projects.list and issues.list, then generate summaries or release notes with aggregated data.
Conversational issue management: embed this MCP server in an agent that accepts natural-language prompts like “Create an issue for failing CI” and translates them to issues.create calls.
Cross-tool integrations: connect an LLM orchestration platform (e.g., an agent that follows MCP) to Plane without modifying the core Plane API usage in your automation scripts.
Audit and approvals: use the server to gate updates through an approval workflow, recording calls and responses centrally before forwarding to Plane.

Quick Tips

Restrict the API key and limit scopes: use a dedicated API key with only the permissions required by the MCP server.
Log and monitor requests: enable logging (LOG_LEVEL) and inspect incoming MCP tool calls to validate LLM behavior.
Validate payloads: when integrating with LLMs, validate and sanitize generated text before using it to create or update issues.

For full source, examples, and contribution guidelines, see the repository on GitHub: https://github.com/kelvin6365/plane-mcp-server.

Plane MCP Server for Project and Issue Management