Sandbox
@eyalzh/kanban-mcp

MCP server for kanban task memory

This project gives an agent a kanban board it can use as working memory for multi-session tasks. It stores boards and tasks in SQLite, exposes MCP tools for creating and moving tasks, and provides prompts for starting or resuming work. A separate web UI lets you inspect and edit the board by hand.

41 stars11 forksTypeScriptUpdated 1y ago
Who it's for

Builders who want their agent to track plans, tasks, and progress across sessions.

What it delivers

You can resume agent work from a board instead of re-explaining the project each time.

What it does

Kanban task memory

Creates boards, adds tasks, moves tasks, deletes tasks, and reads full board or task details through MCP tools.

Wip limit support

Moves tasks between columns while respecting column capacity and work-in-progress limits.

SQLite storage

Uses an embedded database for board state and task history.

Web UI

Provides a browser interface to watch progress and edit tasks manually.

Project prompts

Includes prompts for creating a board and for resuming work on an existing project.

How to get it

  1. 1Clone this repo, then run
    npm ci --prefix shared/db
    npm ci --prefix mcp-server
    npm run build --prefix shared/db
    npm run build --prefix mcp-server
  2. 2Build the Docker image
    docker build -t mcp/mcp-kanban .

README

MCP Kanban Task Management

An MCP tool set providing kanban-based task management state for AI-driven development. This is similar to other task management MCP tools, with the additional structure, rules and visibility of a kanban-based task management system.

The main idea is to direct the AI agent to document and save its work as tasks in a kanban board, both in the planning session and in execution sessions.

Highlights

  • Column capacity / work-in-progress limits
  • Embedded DB (SQLite)
  • Web UI for observing the progress of the workflow, and for modifying tasks manually
  • Predefined prompts for starting and resuming a workflow

Usage

Use the MCP prompts to start a project or to make progress on a project. Alternatively, ask the LLM assistant to record its plan by creating a kanban board. To make progress in a follow-up session, ask the assistant to locate a specific kanban board and resume work on it.

Installation

Clone this repo, then run:

npm ci --prefix shared/db
npm ci --prefix mcp-server
npm run build --prefix shared/db
npm run build --prefix mcp-server

Then add the MCP server configuration to the MCP client (e.g. Claude Desktop):

"mcpServers": {
    "kanban-mcp": {
        "command": "node",
        "args": [
            "/path/to/repo/mcp-server/dist/server.js"
        ],
        "env": {
            "MCP_KANBAN_DB_FOLDER_PATH": "/path/to/db"
        }
    }
}

Replace /path/to/repo with the location of the cloned repo.

Replace /path/to/db with a folder that will contain the DB files. This can be any folder with read/write access. It will be created if it doesn't exist.

Installation with Docker

Build the Docker image:

docker build -t mcp/mcp-kanban .

Then add the MCP server configuration to the MCP client (e.g. Claude Desktop):

"mcpServers": {
    "kanban-mcp": {
        "command": "docker",
        "args": [
            "run",
            "--rm",
            "-i",
            "-v",
            "/path/to/db:/mcp",
            "mcp/mcp-kanban"
        ]
    }
}

Replace /path/to/db with a folder that will contain the DB files. This can be any folder with read/write access. It will be created if it doesn't exist.

Using the web UI

Build the web-ui and web-server:

npm ci --prefix shared/db
npm run build --prefix shared/db

npm ci --prefix web-ui
npm ci --prefix web-server
npm run build --prefix web-ui
npm run build --prefix web-server

Run the web-server:

MCP_KANBAN_DB_FOLDER_PATH=/path/to/db npm run start --prefix web-server

Open the browser at http://localhost:8221

API

Tools

  • create-kanban-board

    • Create a new kanban board to plan and keep track of your tasks.
    • Input:
      • name (string): The name of the board
      • projectGoal (string): The goal of the project.
  • add-task-to-board

    • Add a new task to the landing column (to-do) of a kanban board.
    • Input:
      • boardId (string): The ID of the board to add the task to
      • title (string): The title of the task
      • content (string): The content of the task in markdown format
  • move-task

    • Move a task from one column to another, respecting WIP limits.
    • Input:
      • taskId (string): The ID of the task to move
      • targetColumnId (string): The ID of the column to move the task to
      • reason (string, optional): The reason for moving the task
  • delete-task

    • Delete a task.
    • Input:
      • taskId (string): The ID of the task to delete
  • get-board-info

    • Get the full info of a kanban board, including columns and tasks.
    • Input:
      • boardId (string): The ID of the board to get info for
    • Return detailed information about the board, including columns and tasks.
  • get-task-info

    • Get the full info of a task, including its content.
    • Input:
      • taskId (string): The ID of the task to get info for
    • Return detailed information about the task, including its content.
  • list-boards

    • List all kanban boards in the database.
    • Input: None
    • Return a list of all boards in the database with their names, creation times, and goals.

Prompts

  • create-kanban-based-project

    • Create a kanban board for a project, ask questions to divide the project into tasks, and add them to the board.
    • Input:
      • description (string): The description of the project
  • make-progress-on-a-project

    • Make progress on an existing project by locating its kanban board, selecting the next task, and working on it.
    • Input:
      • description (string): The description of the project

Files in the repo

Repository payload9 top-level entries
  • mcp-server
  • shared
  • web-server
  • web-ui
  • .dockerignore
  • .gitignore
  • Dockerfile
  • LICENSE
  • README.md

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More connectors

Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface

86k

High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.

43k

Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama…) with Codex CLI, App, SDK, and Claude Code

14k
okf-memory/
okf-agent-memory

Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.

547
tirth8205/
code-review-graph

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.

31k
2akouwu/
reverify

Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.

1.1k