About π Kanban MCP
Kanban MCP is a middleware layer that connects the Planka kanban application to LLMs, letting AI assistants manage projects, boards, lists and cards. It targets developers and teams who use Planka and want AI help organizing tasks.
Key features
- 01Manage projects, boards, lists and cards
- 02Create, update and delete cards
- 03Move and duplicate cards
- 04Manage tasks and comments within cards
- 05Start, stop and reset time tracking on tasks
Use cases
- 01Letting an AI assistant organize and update development tasks
- 02Running a workflow where an LLM implements tasks and humans review them
- 03Using AI to plan and groom tasks on a board
Configuration
Local server config is the same for every client. Copy it into your clientβs MCP config file.
We havenβt collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEπ Kanban MCP
Welcome to the Kanban MCP project! π This project integrates Planka kanban boards with Cursor's Machine Control Protocol (MCP) to enable AI assistants like Claude to manage your kanban boards.
π€ What is Kanban MCP?
Kanban MCP is a bridge between Planka (an open-source kanban board) and Cursor's Machine Control Protocol. It allows AI assistants like Claude to:
- π View and manage projects, boards, lists, and cards
- β Create and update tasks
- π¬ Add comments to cards
- π Move cards between lists
- β±οΈ Track time spent on tasks
- π And much more!
This integration enables a seamless workflow where you can ask Claude to help manage your development tasks, track progress, and organize your work.
π¦ Quick Start
π Prerequisites
- π³ Docker for running Planka
- π Git for cloning the repository
- π’ Node.js (version 18 or above) and npm for development
π₯ Installation
- Clone this repository:
git clone https://github.com/bradrisse/kanban-mcp.git
cd kanban-mcp
- Install dependencies and build the TypeScript code:
npm install
npm run build
- Start the Planka containers:
npm run up
-
Access the Planka Kanban board:
- Default URL: http://localhost:3333
- Default credentials:
- Email: [email protected]
- Password: demo
-
Configure Cursor to use the MCP server:
- In Cursor, go to Settings > Features > MCP
- Add a new MCP server with the following configuration:
{ "mcpServers": { "kanban": { "command": "node", "args": ["/path/to/kanban-mcp/dist/index.js"], "env": { "PLANKA_BASE_URL": "http://localhost:3333", "PLANKA_AGENT_EMAIL": "[email protected]", "PLANKA_AGENT_PASSWORD": "demo" } } } }- Replace
/path/to/kanban-mcpwith the actual absolute path to your kanban-mcp directory
Alternatively, you can use a project-specific configuration by creating a .cursor/mcp.json file in your project root with the same configuration.
For Docker-based deployment and other advanced options, see the Installation Guide.
π Documentation
For Users
- π οΈ Installation Guide: How to install and configure Kanban MCP
- π Usage Guide: How to use Kanban MCP with Claude
- π‘ Capabilities and Strategies: Detailed exploration of MCP server capabilities and LLM interaction strategies
- β οΈ Troubleshooting: Solutions to common issues
For Developers
- π¨βπ» Developer Guide: Information for developers who want to contribute to or modify Kanban MCP
- π API Reference: Detailed documentation of the MCP commands and Planka API integration
β¨ Features
Kanban MCP provides a comprehensive set of features for managing your kanban boards:
π Project Management
- Create and view projects
- Manage project settings and members
π Board Management
- Create and view boards within projects
- Customize board settings
π List Management
- Create and organize lists within boards
- Reorder lists as needed
ποΈ Card Management
- Create, update, and delete cards
- Move cards between lists
- Add descriptions, due dates, and labels
- Duplicate cards to create templates
β±οΈ Time Tracking
- Start, stop, and reset stopwatches
- Track time spent on individual tasks
- Analyze time usage patterns
β Task Management
- Create and manage tasks within cards
- Mark tasks as complete or incomplete
π¬ Comment Management
- Add comments to cards for discussion
- View comment history
π€ LLM Interaction Strategies
MCP Kanban supports several workflow strategies for LLM-human collaboration:
- π€ LLM-Driven Development with Human Review: LLMs implement tasks while humans review and provide feedback
- π₯ Human-Driven Development with LLM Support: Humans implement while LLMs provide analysis and recommendations
- π§ Collaborative Grooming and Planning: Humans and LLMs work together to plan and organize tasks
For more details on these strategies, see the Capabilities and Strategies wiki page.
π¦ Available npm Scripts
npm run build: Build the TypeScript codenpm run build-docker: Build the TypeScript code and create a Docker imagenpm run up: Start the Planka containers (kanban and postgres)npm run down: Stop all containersnpm run restart: Restart the Planka containersnpm run start-node: Start the MCP server directly with Node (for testing outside of Cursor)npm run qc: Run quality control checks (linting and type checking)
π€ Contributing
We welcome contributions to Kanban MCP! If you'd like to contribute:
- Check out the Developer Guide for information on the project structure and development workflow
- Look at the open issues for tasks that need help
- Submit a pull request with your changes
π Support
If you need help with Kanban MCP:
- Check the Troubleshooting page for solutions to common issues
- Search the GitHub issues to see if your problem has been reported
- Open a new issue if you can't find a solution
π License
Kanban MCP is open-source software licensed under the MIT License. See the LICENSE file for details.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.