💡 Summary
Claude MPM is a multi-agent orchestration framework for managing workflows and monitoring in Claude Code CLI.
🎯 Target Audience
🤖 AI Roast: “It's like a Swiss Army knife, but for code—just don't forget the batteries.”
The README suggests potential risks such as dependency supply chain issues and environment variable exposure (GITHUB_TOKEN). To mitigate, ensure secure storage of tokens and regularly audit dependencies.
Claude MPM - Multi-Agent Project Manager
A powerful orchestration framework for Claude Code (CLI) that enables multi-agent workflows, session management, and real-time monitoring through a streamlined Rich-based interface.
⚠️ Important: Claude MPM requires Claude Code CLI (v2.1.3+), not Claude Desktop (app). All MCP integrations work with Claude Code's CLI interface only.
Don't have Claude Code? Install from: https://docs.anthropic.com/en/docs/claude-code
Quick Start: See Getting Started Guide to get running in 5 minutes!
Who Should Use Claude MPM?
- 👥 Non-Technical Users (Founders/PMs) - Research and understand codebases using Research Mode - no coding experience required
- 💻 Developers - Multi-agent development workflows with semantic code search and advanced features
- 🏢 Teams - Collaboration patterns, session management, and coordinated workflows
What is Claude MPM?
Claude MPM transforms Claude Code into a multi-agent orchestration platform with:
- 47+ Specialized Agents - From Git repositories (Python, Rust, QA, Security, Ops, etc.)
- Intelligent Task Routing - PM agent delegates work to specialist agents
- Session Management - Resume previous sessions with full context preservation
- Semantic Code Search - AI-powered discovery of existing code and patterns
- Real-Time Monitoring - Live dashboard showing agent activity and performance
- Git Repository Integration - Always up-to-date agents and skills from curated repositories
Quick Installation
Prerequisites
- Python 3.11-3.13 (Python 3.13 recommended; 3.14 NOT yet supported)
- Claude Code CLI v2.1.3+ (required!)
- GitHub Token (recommended for skill sources)
Python Version Warning:
- macOS default Python 3.9 is too old - use
--python 3.13flag- Python 3.13 is recommended and fully tested
- Python 3.14 is NOT yet supported - installation will fail
# Verify Claude Code is installed claude --version # If not installed, get it from: # https://docs.anthropic.com/en/docs/claude-code # Set GitHub token (recommended - avoids rate limits) export GITHUB_TOKEN=your_github_token
Install Claude MPM
IMPORTANT: Install from your home directory, NOT from within a cloned git repository.
uv (recommended):
# From home directory (IMPORTANT!) cd ~ # Install with Python 3.13 (not 3.9 or 3.14) uv tool install claude-mpm[monitor,data-processing] --python 3.13
Homebrew (macOS):
brew tap bobmatnyc/tools brew install claude-mpm
pipx:
cd ~ pipx install "claude-mpm[monitor]"
Post-Installation Setup (Required)
These steps must be completed before running claude-mpm doctor:
# Create required directories mkdir -p ~/.claude/{responses,memory,logs} # Deploy agents claude-mpm agents deploy # Add skill source (recommended) claude-mpm skill-source add https://github.com/bobmatnyc/claude-mpm-skills
Verify Installation
# Run diagnostics (after completing setup above) claude-mpm doctor --verbose # Check versions claude-mpm --version claude --version # Auto-configure your project cd ~/your-project claude-mpm auto-configure
What You Should See:
- 47+ agents deployed to
~/.claude/agents/ - 17 bundled skills (in Python package)
- Agent sources configured
- All doctor checks passing
Recommended Partners: Install these companion tools for enhanced capabilities:
uv tool install kuzu-memory --python 3.13 uv tool install mcp-vector-search --python 3.13 uv tool install mcp-ticketer --python 3.13 uv tool install mcp-browser --python 3.13
Tool Version Management: Use ASDF version manager to avoid Python/uv version conflicts across projects.
Key Features
🎯 Multi-Agent Orchestration
- 47+ Specialized Agents from Git repositories covering all development needs
- Smart Task Routing via PM agent intelligently delegating to specialists
- Session Management with
--resumeflag for seamless continuity - Resume Log System with automatic 10k-token summaries at 70%/85%/95% thresholds
→ Learn more: Multi-Agent Development
📦 Git Repository Integration
- Curated Content with 47+ agents automatically deployed from repositories
- Always Up-to-Date with ETag-based caching (95%+ bandwidth reduction)
- Hierarchical BASE-AGENT.md for template inheritance and DRY principles
- Custom Repositories via
claude-mpm agent-source add
🎯 Skills System
- 17 Bundled Skills covering Git, TDD, Docker, API docs, testing, and more
- Three-Tier Organization: Bundled/user/project with priority resolution
- Auto-Linking to relevant agents based on roles
- Custom Skills via
.claude/skills/or skill repositories
🔍 Semantic Code Search
- AI-Powered Discovery with mcp-vector-search integration
- Find by Intent not just keywords ("authentication logic" finds relevant code)
- Pattern Recognition for discovering similar implementations
- Live Updates tracking code changes automatically
→ Learn more: Developer Use Cases
🧪 MPM Commander (ALPHA)
- Multi-Project Orchestration with autonomous AI coordination across codebases
- Tmux Integration for isolated project environments and session management
- Event-Driven Architecture with inbox system for cross-project communication
- LLM-Powered Decisions via OpenRouter for autonomous work queue processing
- Real-Time Monitoring with state tracking (IDLE, WORKING, BLOCKED, PAUSED, ERROR)
- ⚠️ Experimental - API and CLI interface subject to change
🔌 Advanced Integration
- MCP Integration with full Model Context Protocol support
- Real-Time Monitoring via
--monitorflag and web dashboard - Multi-Project Support with per-session working directories
- Git Integration with diff viewing and change tracking
🔐 OAuth & Google Workspace Integration
- Browser-Based OAuth for secure authentication with MCP services
- Google Workspace MCP built-in server for Gmail, Calendar, and Drive
- Encrypted Token Storage using Fernet encryption with system keychain
- Automatic Token Refresh handles expiration seamlessly
# Set up Google Workspace OAuth claude-mpm oauth setup workspace-mcp # Check token status claude-mpm oauth status workspace-mcp # List OAuth-capable services claude-mpm oauth list
→ Learn more: OAuth Setup Guide
⚡ Performance & Security
- Simplified Architecture with ~3,700 lines removed for better performance
- Enhanced Security with comprehensive input validation
- Intelligent Caching with ~200ms faster startup via hash-based invalidation
- Memory Management with cleanup commands for large conversation histories
⚙️ Automatic Migrations
- Seamless Updates with automatic configuration migration on first startup after update
- One-Time Fixes for cache restructuring and configuration changes
- Non-Blocking failures log warnings but do not stop startup
- Tracked in
~/.claude-mpm/migrations.yaml
→ Learn more: Startup Migrations
Quick Usage
# Start interactive mode claude-mpm # Start with monitoring dashboard claude-mpm run --monitor # Resume previous session claude-mpm run --resume # Semantic code search claude-mpm search "authentication logic" # or inside Claude Code: /mpm-search "authentication logic" # Health diagnostics claude-mpm doctor # Verify MCP services claude-mpm verify # Manage memory claude-mpm cleanup-memory
💡 Update Checking: Claude MPM automatically checks for updates and verifies Claude Code compatibility on startup. Configure in ~/.claude-mpm/configuration.yaml or see docs/update-checking.md.
→ Complete usage examples: User Guide
What's New in v5.0
Git Repository Integration for Agents & Skills
- 📦 Massive Library: 47+ agents and hundreds of skills deployed automatically
- 🏢 Official Content: Anthropic's official skills repository included by default
- 🔧 Fully Extensible: Add your own repositories with immediate testing
- 🌳 Smart Organization: Hierarchical BASE-AGENT.md inheritance
- 📊 Clear Visibility: Two-phase progress bars (sync + deployment)
- ✅ Fail-Fast Testing: Test repositories before they cause startup issues
Quick Start with Custom Repositories:
# Add custom agent repository claude-mpm agent-source add https://github.com/yourorg/your-agents # Add custom skill repository claude-mpm skill-source add https://github.com/yourorg/your-skills # Test repository without saving claude-mpm agent-source add https://github.com/yourorg/your-agents --test
Documentation
📚 Complete Documentation Hub - Start here for all documentation!
Quick Links by User Type
👥 For Users
- 🚀 5-Minute Quick Start - Get running immediately
- 📦 Installation Guide - All installation methods
- 📖 User Guide - Complete user documentation
- ❓ FAQ - Common questions answered
💻 For Devel
Pros
- Supports multi-agent workflows.
- Real-time monitoring capabilities.
- Easy integration with Git repositories.
- User-friendly for non-technical users.
Cons
- Requires specific Python version.
- Installation can be complex for beginners.
- Limited to Claude Code CLI.
- Some features are still in alpha.
Related Skills
pytorch
S“It's the Swiss Army knife of deep learning, but good luck figuring out which of the 47 installation methods is the one that won't break your system.”
agno
S“It promises to be the Kubernetes for agents, but let's see if developers have the patience to learn yet another orchestration layer.”
nuxt-skills
S“It's essentially a well-organized cheat sheet that turns your AI assistant into a Nuxt framework parrot.”
Disclaimer: This content is sourced from GitHub open source projects for display and rating purposes only.
Copyright belongs to the original author bobmatnyc.
