Skillget

planka-v2-mcp

mcp
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 7 GitHub stars
Code Uyari
  • process.env — Environment variable access in .jest/setEnvVars.example.js
  • process.env — Environment variable access in common/setup.ts
  • process.env — Environment variable access in common/utils.ts
  • network request — Outbound network request in common/utils.ts
  • process.env — Environment variable access in index.ts
Permissions Gecti
  • Permissions — No dangerous permissions requested

Bu listing icin henuz AI raporu yok.

SUMMARY

Official Model Context Protocol (MCP) server for Planka v2.x Kanban. Supports Stdio & HTTP SSE, Docker, and AI workflow automation with Cursor & Claude.

README.md

📋 Planka Kanban MCP (v2.x)

A specialized Model Context Protocol (MCP) server that enables LLMs (like Claude in Cursor) to interact with Planka v2.x kanban boards.

[!IMPORTANT]
This MCP server is designed specifically for Planka v2.x. It allows an AI agent to manage projects, boards, cards, and the new Task List hierarchy on an existing Planka instance.

✨ Key Features (Planka v2.x)

  • 📂 Task List Management: Full support for the new Task List entity introduced in Planka v2.0.
  • ✅ Granular Task Tracking: Create and complete tasks within specific Task Lists inside cards.
  • ⏱️ Persistent Time Tracking: Start and stop stopwatches on cards; time is tracked directly in Planka.
  • 👥 Unified Membership Management: Assign users to cards by Email or Username without needing their ID first.
  • 🚀 Project-First Defaults: New cards are created as type "project" by default to ensure subtasks and checklists are immediately visible.
  • 🗂️ Consolidated Managers: Grouped tools (Card Manager, Project Manager, etc.) to optimize LLM interaction and stay under tool limits.
  • ⚡ Workflow Automation: Built-in workflow_action on mcp_kanban_card_manager allowing single-call card status progression (start_working, mark_completed, move_to_testing, move_to_done) with optional comments and task completion.
  • 📊 Aggregate Summaries: New tools for comprehensive data retrieval:
    Tool Name Parameters Description
    get_board id Get board details.
    update_board id, name, backgroundType, etc. Update board settings.
    delete_board id Delete a board.
    get_board_summary boardId, includeTaskDetails, includeComments Get a comprehensive summary of a board including stats.
    get_project_summary id Get a summary of all boards in a project.
  • 🏷️ Advanced Labeling: Full support for v2.x color palettes and label assignment.

🚀 Quick Start (Connect to your Planka v2.x)

1. Prerequisites

  • A functional Planka v2.x instance reachable via network.
  • An agent user created in Planka with the necessary permissions.

2. Configure Your Client (Cursor/Claude/Antigravity)

The easiest way is using npx. Add the following to your MCP configuration:

{
  "mcpServers": {
    "planka-mcp": {
      "command": "npx",
      "args": ["-y", "@navyatec/planka-v2-mcp@latest"],
      "env": {
        "PLANKA_BASE_URL": "https://your-planka-instance.com",
        "PLANKA_AGENT_EMAIL": "[email protected]",
        "PLANKA_AGENT_PASSWORD": "your-secure-password",
        "PLANKA_IGNORE_SSL": "true"
      }
    }
  }
}

Alternative: Local Setup

If you have the code locally, pointing to the built file:

{
  "mcpServers": {
    "planka-mcp": {
      "command": "node",
      "args": ["/absolute/path/to/kanban-mcp/dist/index.js"],
      "env": { ... }
    }
  }
}

🌐 HTTP SSE Mode (Remote / Cloud Agents)

You can run the server in HTTP Server-Sent Events (SSE) mode to connect remote clients or centralize deployment:

# Start directly via CLI
npx -y @navyatec/planka-v2-mcp --transport sse --port 3000

# Or using environment variables
export MCP_TRANSPORT=sse
export PORT=3000
export MCP_API_KEY=your-optional-bearer-key
node dist/index.js

SSE Endpoints:

  • GET http://localhost:3000/sse: Persistent SSE stream handshake (?token=... supported)
  • POST http://localhost:3000/messages?sessionId=...: Client JSON-RPC requests
  • GET http://localhost:3000/health: Service healthcheck

🐳 Docker & GHCR

Run directly from GitHub Container Registry:

docker run -d -p 3000:3000 \
  -e PLANKA_BASE_URL="http://your-planka-instance:3333" \
  -e PLANKA_AGENT_EMAIL="[email protected]" \
  -e PLANKA_AGENT_PASSWORD="your-secure-password" \
  -e MCP_API_KEY="optional-mcp-bearer-token" \
  ghcr.io/navya-tecnologia/planka-v2-mcp:latest

Ask Claude:

  • "Groom the Backlog board and break down the top card into tasks."
  • "Start tracking time for the 'UI Implementation' card."
  • "Assign @jane.doe to the card 'Implement API Authentication'."
  • "Create a new 'Database' Task List in the 'System Migration' card."

📖 Documentation

🏗️ Development

# Install dependencies
npm install

# Build the project
npm run build

# Run unit tests (offline, in-memory)
npm test

# Run integration tests (requires running Planka instance)
npm run test:integration

# Run quality checks (build + unit tests)
npm run qc

🚀 Automated Releases

This project is configured for automated releases via GitHub Actions. Pushing a tag in the format v* (e.g., v1.4.0) will automatically run tests, build, publish the package to NPM, publish the container image to GHCR, and create a GitHub Release.

This project focuses exclusively on the MCP interface. For Planka server setup, refer to the official Planka documentation.
Developed by NAVYA TECNOLOGÍA 2024, S.L.

Yorumlar (0)

Sonuc bulunamadi