cookjohn

cookjohn / teammcp

Public

MCP-native collaboration server for AI agent teams — real-time messaging, task management, and web dashboard with just 1 npm dependency

14
0
100% credibility
Found Mar 31, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

TeamMCP is a lightweight collaboration server that lets multiple AI agents communicate via group chats, direct messages, task tracking, full-text search, and a real-time web dashboard.

How It Works

1
🔍 Discover TeamMCP

You hear about TeamMCP, a fun way to let your AI helpers chat and work together like a real team in an office.

2
📥 Get it ready

Download and run the simple setup so everything is prepared on your computer in just a few moments.

3
🚀 Launch the team hub

Turn on the central spot where all your AI friends can connect and start collaborating right away.

4
👥 Add team members

Give names and roles to your AI helpers, like engineer or manager, so they know who they are.

5
🔗 Connect your AIs

Link each AI helper to the team hub so they can send messages and see updates.

6
📱 Open the dashboard

Visit the web page to watch live chats, check tasks, and see your team buzzing with activity.

🎉 Team magic happens

Your AI team now talks in channels, shares tasks, searches ideas, and builds projects together seamlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is teammcp?

TeamMCP is a JavaScript MCP-native collaboration server for AI agent teams, delivering real-time messaging, task management, and a web dashboard with just one npm dependency. It lets persistent AI agents—running in Claude Code—chat via group channels or DMs, assign tasks, search message history, and monitor status through a browser interface. Instead of rigid orchestration, it enables natural agent-to-agent communication over the open MCP protocol, backed by SQLite for lightweight persistence.

Why is it gaining traction?

It ditches heavy frameworks like CrewAI or LangGraph for a single-package setup, offering 20 MCP tools for sending messages, creating tasks, and pulling inboxes, plus a real-time SSE stream and full-text search under 100ms. Developers hook it up fast via curl registration and Claude's mcp add command, with seamless integration into AgentRegistry for discovery and AgentGateway for secure routing. The web dashboard gives instant visibility into agent teams without extra tooling.

Who should use this?

AI developers building multi-agent workflows in Claude Code who need a communication layer beyond scripted chats. Experimenters simulating engineering teams—think CEO delegating to engineers via tasks and channels—or researchers testing persistent agent collaboration. It's ideal for JavaScript devs prototyping agent teams without dependency bloat.

Verdict

Try TeamMCP if you're into MCP-based agent experiments; its docs and quickstart shine for a 14-star project, but the 1.0% credibility score signals early maturity—expect to tweak for production. Solid for proofs-of-concept, pair with AgentGateway for real teams.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.