# pkanban

> pkanban is an MCP server for productivity. A kanban board an AI reads and updates as it works, which you watch in your browser. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** CLI, Web, API, self-hosted, macOS, Linux, Windows
- **Website:** https://pkanban.pearachute.com/docs/mcp
- **Repository:** https://github.com/japherwocky/pkanban
- **MCP Registry name:** io.github.japherwocky/pkanban
- **Categories:** [productivity](https://ai-powered.com/categories/productivity), [developer tools](https://ai-powered.com/categories/developer-tools)
- **Listing updated:** 2026-11-18

## About

pkanban is a command-line-first kanban board built for people and AI agents together, and this MCP server is the route for AI apps that cannot run shell commands, such as the Claude and ChatGPT chat apps. The same boards, columns and cards are reachable from the CLI, the hosted site and the MCP endpoint at `https://pkanban.pearachute.com/mcp`.

The CLI covers boards, columns and cards, and sharing a board with a team at read, write or admin level. For automation it supports revocable API keys that are stored bcrypt-hashed and track last use, and a `--json` flag prints raw API responses so scripts avoid parsing prose. Login opens an approval page in the browser, so no secret is typed into a terminal and an agent can be signed in the same way.

Most users use the hosted service, and the README includes a self-hosting section.

## Features

- **Boards, columns, cards:** Create and update kanban structure from a CLI or an AI app.
- **Team sharing:** Share boards with read, write or admin permission.
- **Browser approval login:** Sign in a terminal or agent by approving in the browser.
- **Revocable API keys:** Keys can be revoked and reactivated, with last-use tracking.
- **JSON output:** `--json` returns raw responses for scripting.
- **Self-hosting:** Run your own server and point the CLI at it.

## Good for

- Let an AI track its own tasks while it works
- Watch an agent's progress on a board in your browser
- Share a project board with a team
- Script card creation in CI with an API key
- Use boards from the Claude or ChatGPT chat apps

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http pkanban https://pkanban.pearachute.com/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://pkanban.pearachute.com/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "pkanban": {
      "url": "https://pkanban.pearachute.com/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "pkanban": {
      "type": "http",
      "url": "https://pkanban.pearachute.com/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "pkanban": {
      "serverUrl": "https://pkanban.pearachute.com/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "pkanban": {
      "type": "http",
      "url": "https://pkanban.pearachute.com/mcp"
    }
  }
}
```

## FAQ

### Can I self-host it?

Yes. The README has a self-hosting section and a command to point the CLI at your own server.

### How do agents sign in?

Run `pkanban login --no-wait`, approve in the browser, then run `pkanban login` again, or use an API key.

### How is it installed?

With a curl or PowerShell installer, or `pip install pkanban`.

### Why an MCP server?

So AI apps that cannot run commands can use the same boards.

### How do I add pkanban to Claude Code?

Run this command in your terminal: claude mcp add --transport http pkanban https://pkanban.pearachute.com/mcp

### What are the best alternatives to pkanban?

The closest alternatives to pkanban are REA, Context7, Chrome DevTools MCP, Codebase Memory and GitHub.

## Where this comes from

- Description: written from the project's README and the project's website.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Context7](https://ai-powered.com/mcp/context7): Context7 is a freemium MCP server for developer tools.
- [Chrome DevTools MCP](https://ai-powered.com/mcp/chrome-devtools-mcp-2): Chrome DevTools MCP is an open-source MCP server for browser automation.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [awesome-mcp-servers](https://ai-powered.com/mcp/awesome-mcp-servers-2): awesome-mcp-servers is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/pkanban
