Skip to main content
Glama

English | 한국어 | 日本語 | 中文

MonkeyPlanner

Local-first task memory for your AI coding agents. Approve with a click; your agents do the rest. No cloud. No telemetry. Forever free, forever MIT.

Works with Claude Code · Claude Desktop · Cursor · Continue · any MCP-compatible client.

MonkeyPlanner Demo

Quickstart

# Docker (recommended)
docker run -p 8080:8080 -v $(pwd)/data:/data ghcr.io/kjm99d/monkeyplanner:latest

# then wire up your agent
monkey-planner mcp install --for claude-code     # or --for cursor / --for claude-desktop

Open http://localhost:8080 — the built-in Welcome board walks you through the rest.

Related MCP server: sprinter

Features

Issue & Board Management

  • Kanban Board — Drag and drop, horizontal scroll, filtering, sorting, and table view toggle

  • Issue Creation — Title, markdown body, and custom properties

  • Custom Properties — Six supported types:

    • Text

    • Number

    • Select

    • Multi-select

    • Date

    • Checkbox

Approval Gate

  • Pending → Approved via a dedicated approval endpoint (cannot be done via generic PATCH)

  • Approval Queue — Bulk-approve all Pending issues across boards

  • Approved → InProgress → Done — Flexible status transitions

  • Rejected status — Record a rejection reason

Agent Features

  • Agent Instructions field — Provide detailed instructions for MCP agents to follow

  • Success Criteria — Manage completion conditions as a checklist

  • Comments — Log progress and communicate per issue

  • Dependencies — Express blocking relationships between issues

Data Visualization

  • Calendar — Monthly grid + daily activity (created, approved, completed counts)

  • Dashboard — Stats cards + weekly activity chart

  • Sidebar — Board list, issue counts, and recent items

User Experience

  • Global Search — Quick search with Cmd+K

  • Keyboard Shortcuts

    • h — Go to dashboard

    • a — Go to approval queue

    • ? — Show shortcut help

    • Cmd+S — Save

    • Escape — Close modal/dialog

  • Collapsible Sidebar — Maximize screen space

  • Dark Mode — Theme toggle

  • Internationalization — Korean, English, Japanese, and Chinese

Automation & Integrations

  • Webhooks — Discord, Slack, and Telegram support

    • Events: issue.created, issue.approved, issue.status_changed, issue.updated, issue.deleted, comment.created

  • Real-time UI sync (SSE) — Changes via MCP/CLI automatically reflect in open browser tabs, no refresh needed

  • JSON Export — Export all issue data

  • Right-click Context Menu — Quick actions

  • Issue Templates — Per-board localStorage persistence

MCP Server (AI Agent Integration)

Thirteen tools for AI agent automation:

  1. list_boards — List all boards

  2. list_issues — Query issues (filter by boardId, status)

  3. get_issue — Issue detail including instructions, criteria, and comments

  4. create_issue — Create a new issue

  5. approve_issue — Approve: Pending → Approved

  6. claim_issue — Claim: Approved → InProgress

  7. submit_qa — Submit for QA: InProgress → QA

  8. complete_issue — Complete: QA → Done (optional comment)

  9. reject_issue — Reject: QA → InProgress with required reason

  10. add_comment — Add a comment to an issue

  11. update_criteria — Check or uncheck a success criterion

  12. search_issues — Search issues by title

  13. get_version — Get the MCP server version (for diagnostics)

Tech Stack

Backend

  • Language: Go 1.26

  • Router: chi/v5

  • Database: SQLite / PostgreSQL (configurable)

  • Migrations: goose/v3

  • Embedded files: embed.FS (single-binary deployment)

Frontend

  • Framework: React 18

  • Language: TypeScript

  • Bundler: Vite 6

  • CSS: Tailwind CSS

  • State management: React Query (TanStack)

  • Drag and drop: @dnd-kit/core, @dnd-kit/sortable

  • Icons: lucide-react

  • Charts: recharts

  • i18n: react-i18next

  • Markdown: react-markdown + rehype-sanitize

MCP

  • Protocol: JSON-RPC 2.0 over stdio

  • Targets: Claude Code, Claude Desktop

Getting Started

Requirements

  • Go 1.26 or later

  • Node.js 18 or later

  • npm or yarn

Installation & Running

1. Clone and initialize

git clone https://github.com/kjm99d/MonkeyPlanner.git
cd monkey-planner
make init

2. Production build (single binary)

make build
./bin/monkey-planner

The server runs at http://localhost:8080 with the frontend embedded.

3. Development mode (separate processes)

Terminal 1 — backend:

make run-backend

Terminal 2 — frontend (Vite dev server, :5173):

make run-frontend

The frontend automatically proxies /api requests to :8080.

Environment Variables

# Server address (default: :8080)
export MP_ADDR=":8080"

# Database connection string
# SQLite (default: sqlite://./data/monkey.db)
export MP_DSN="sqlite://./data/monkey.db"

# PostgreSQL example
export MP_DSN="postgres://user:password@localhost:5432/monkey_planner"

MCP Server Setup

# Claude Code (writes .mcp.json in the current directory)
monkey-planner mcp install --for claude-code

# Claude Desktop (writes the OS-native config file)
monkey-planner mcp install --for claude-desktop

# Cursor (writes .cursor/mcp.json)
monkey-planner mcp install --for cursor

Flags: --dry-run to preview, --scope user for a global entry (~/.mcp.json), --force to overwrite, --base-url <url> to point at a non-default server.

Restart the client afterwards so it re-reads the config.

Manual configuration

Works identically for Claude Code (.mcp.json), Claude Desktop (OS-native config), and Cursor (.cursor/mcp.json):

{
  "mcpServers": {
    "monkey-planner": {
      "command": "/path/to/monkey-planner",
      "args": ["mcp"],
      "env": {
        "MP_BASE_URL": "http://localhost:8080"
      }
    }
  }
}

The binary must be able to reach the HTTP server (set with MP_BASE_URL). Leave it at the default when running both on the same machine.

MCP Tool Usage Examples

AI: List all boards
→ list_boards()

AI: Find issues related to "authentication"
→ search_issues(query="authentication")

AI: Approve the first pending issue, claim it, work on it, and submit for QA
→ approve_issue() → claim_issue() → submit_qa()

Workflow — Real Usage Scenario

Below is a real workflow from fixing a language switcher bug, showing how a human and AI agent collaborate through MonkeyPlanner.

Status Flow

Pending → Approved → InProgress → QA → Done
                         ↑              │ (reject with reason)
                         └──────────────┘

Step-by-Step

1. Create Issue — Human finds a bug, asks AI to register it

Human: "The language selector dropdown doesn't appear when clicking the button. Create an issue."
AI:    create_issue(boardId, title, body, instructions)  →  status: Pending

2. Approve — Human reviews and approves

Human: (clicks Approve on the board or tells AI)
AI:    approve_issue(issueId)  →  status: Approved

3. Start Work — AI claims the issue and begins coding

AI:    claim_issue(issueId)  →  status: InProgress
       - Reads code, identifies root cause
       - Implements fix, runs tests
       - Commits changes

4. Submit for QA — AI finishes and submits for review

AI:    submit_qa(issueId, comment: "commit abc1234 — fixed click handler")
       →  status: QA
       add_comment(issueId, "Commit info: ...")

5. Review — Human tests the fix

Human: Tests in browser, finds the dropdown is clipped by sidebar
       →  reject_issue(issueId, reason: "Dropdown is hidden behind sidebar")
       →  status: InProgress  (back to step 3)

Human: Tests again after fix, everything works
       →  complete_issue(issueId)  →  status: Done

6. Feedback Loop — Communication via comments throughout

Human: add_comment("Dropdown is clipped on the left side, fix it")
AI:    get_issue() → reads comment → fixes → commit → submit_qa()
Human: Tests → complete_issue()  →  Done ✓

Key Takeaways

  • Human controls the gates: Approve, QA pass/reject, Complete

  • AI does the work: Code analysis, implementation, testing, commits

  • Comments are the communication channel: Both sides use add_comment to exchange feedback

  • QA loop prevents premature completion: Issues must pass human review before Done

API Reference

OpenAPI 3.0 spec: backend/docs/swagger.yaml

Key Endpoints

Boards

GET    /api/boards                  # List boards
POST   /api/boards                  # Create board
PATCH  /api/boards/{id}             # Update board
DELETE /api/boards/{id}             # Delete board

Issues

GET    /api/issues                  # List issues (filter: boardId, status, parentId)
POST   /api/issues                  # Create issue
GET    /api/issues/{id}             # Issue detail + child issues
PATCH  /api/issues/{id}             # Update issue (status, properties, title, etc.)
DELETE /api/issues/{id}             # Delete issue
POST   /api/issues/{id}/approve     # Approve issue (Pending → Approved)

Comments

GET    /api/issues/{issueId}/comments    # List comments
POST   /api/issues/{issueId}/comments    # Add comment
DELETE /api/comments/{commentId}         # Delete comment

Properties (Custom Attributes)

GET    /api/boards/{boardId}/properties      # List property definitions
POST   /api/boards/{boardId}/properties      # Create property
PATCH  /api/boards/{boardId}/properties/{propId}  # Update property
DELETE /api/boards/{boardId}/properties/{propId}  # Delete property

Webhooks

GET    /api/boards/{boardId}/webhooks           # List webhooks
POST   /api/boards/{boardId}/webhooks           # Create webhook
PATCH  /api/boards/{boardId}/webhooks/{whId}    # Update webhook
DELETE /api/boards/{boardId}/webhooks/{whId}    # Delete webhook

Calendar

GET /api/calendar           # Monthly stats (year, month required)
GET /api/calendar/day       # Daily issue list (date required)

For full schema details, see backend/docs/swagger.yaml.

Project Structure

monkey-planner/
├── backend/
│   ├── cmd/monkey-planner/
│   │   ├── main.go              # Entry point (HTTP server)
│   │   └── mcp.go               # MCP server (JSON-RPC stdio)
│   ├── internal/
│   │   ├── domain/              # Domain models (Issue, Board, etc.)
│   │   ├── service/             # Business logic
│   │   ├── storage/             # Database layer (SQLite/PostgreSQL)
│   │   ├── http/                # HTTP handlers & router
│   │   └── migrations/          # goose migration files
│   ├── web/                     # Embedded frontend (embed.FS)
│   ├── docs/
│   │   └── swagger.yaml         # OpenAPI 3.0 spec
│   ├── go.mod
│   └── go.sum
│
├── frontend/
│   ├── src/
│   │   ├── components/          # Reusable components
│   │   ├── features/            # Page & feature components
│   │   │   ├── home/           # Dashboard
│   │   │   ├── board/          # Board & Kanban
│   │   │   ├── issue/          # Issue detail
│   │   │   ├── calendar/       # Calendar
│   │   │   └── approval/       # Approval queue
│   │   ├── api/                 # API hooks & client
│   │   ├── design/              # Tailwind tokens
│   │   ├── i18n/                # Translations (en.json, ko.json, ja.json, zh.json)
│   │   ├── App.tsx              # Router
│   │   ├── index.css            # Global styles
│   │   └── main.tsx
│   ├── package.json
│   ├── vite.config.ts
│   ├── tsconfig.json
│   └── tailwind.config.js
│
├── .mcp.json                    # Claude Code MCP config
├── Makefile                     # Build & dev commands
├── .githooks/                   # Git hooks
└── data/                        # SQLite database (default)

Testing

Backend tests

make test-backend

Frontend tests

make test-frontend

Accessibility tests

make test-a11y

All tests

make test

Common Commands

# Initial setup after cloning
make init

# Production build
make build

# Run production server
./bin/monkey-planner

# Development mode
make run-backend        # Terminal 1
make run-frontend       # Terminal 2

# Clean build artifacts
make clean

Status Transition Rules

Pending
  ↓ (approve endpoint)
Approved
  ↓ (PATCH status)
InProgress
  ↓ (PATCH status)
Done

Pending → Approved: POST /api/issues/{id}/approve (dedicated endpoint only)
Approved ↔ InProgress ↔ Done: Free transitions via PATCH
Pending: Cannot be re-entered from other statuses
Rejected: Separate rejection state with reason tracking

License

MIT

Contributing

Issues and pull requests are welcome.

Contact

For questions or feedback about the project, please open a GitHub Issue.

A
license - permissive license
-
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
0dRelease cycle
9Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • A
    license
    -
    quality
    D
    maintenance
    An MCP server that provides a database-backed kanban board with 40+ tools for AI agents to track issues, features, todos, epics, and diary entries across projects, including status workflows, relationships, and semantic search.
    Last updated
    77
    MIT
  • A
    license
    -
    quality
    C
    maintenance
    A self-hosted backlog tracker with priority scoring and an MCP server, enabling AI agents to autonomously pull, work on, and update tasks via JSON-RPC tools.
    Last updated
    MIT
  • A
    license
    -
    quality
    B
    maintenance
    An AI-native kanban board MCP server where agents pull tasks via the Model Context Protocol. It provides tools for task discovery, claiming, progress updates, and review workflows.
    Last updated
    1
    MIT

View all related MCP servers

Related MCP Connectors

  • Control plane for autonomous software labor. Agents claim objectives over MCP with audit trail.

  • Local-first RAG engine with MCP server for AI agent integration.

  • AI Reasoning Cache & Consensus Layer with 11 MCP tools via Streamable HTTP.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/kjm99d/MonkeyPlanner'

If you have feedback or need assistance with the MCP directory API, please join our Discord server