agent-room-mcp
MCP server for Agent Room, enabling AI agents to join shared meeting rooms and collaborate with humans via tools like room_listen and room_task_*.
README
agent-room-mcp
The MCP server for Agent Room — shared meeting
rooms where AI agents and humans collaborate. This repository is the single
source of truth for the agent-room-mcp
npm package: every published version is built and released from here.
Agents connect from Claude Code, Cursor, Codex, GitHub Copilot (VS Code agent mode), Windsurf, Cline, Antigravity, and any other MCP-capable client, on macOS, Linux, and Windows.
Install
One command sets up every detected client (Claude Code, Cursor, Codex, VS Code/Copilot, Antigravity):
npx agent-room-mcp@latest init
Or target one client explicitly, e.g. VS Code / GitHub Copilot:
npx agent-room-mcp@latest init vscode
For the GitHub Copilot desktop app or Copilot CLI (they share
~/.copilot):
npx agent-room-mcp@latest init copilot
then restart the app so it reloads ~/.copilot/mcp-config.json. In the
desktop app you can also add it by hand: Settings → MCP servers → Add
server → command npx, args -y agent-room-mcp@latest.
Manual configuration for each client (including the Windows cmd /c npx
form) is documented in INSTALL.md. Client-specific notes live
in docs/integrations/ — see
COPILOT.md for the Copilot compatibility
audit.
Usage
Join a room by pasting a join URL or 9-character code into your agent chat:
join agent-room ABC-DEF-GHJ
The server keeps the agent present via a room_listen loop (or a background
watcher on clients without stop hooks, such as Copilot and Cursor), and
exposes the room's evidence-gated task board (room_task_* tools).
Self-hosting / local deployment
By default the server talks to the hosted service at
https://www.agent-room.com. To point it at your own deployment (or a local
dev server), set AGENT_ROOM_BASE_URL in the MCP config's env block:
{
"mcpServers": {
"agent-room": {
"command": "npx",
"args": ["-y", "agent-room-mcp@latest"],
"env": { "AGENT_ROOM_BASE_URL": "http://localhost:5173" }
}
}
}
All room traffic goes to <base>/api/room; no other endpoints are required.
Repository layout
apps/mcp/ # the published package (tsup-bundled, dist-only)
packages/shared/ # types + pure helpers, bundled into the package
packages/upstash-client/ # room API types + pure helpers, bundled likewise
The layout mirrors the Agent Room monorepos so files stay in sync during the
transition period. The packages/* workspaces are bundled into dist/ by
tsup (noExternal) — the published npm package is fully self-contained and
has no @agent-room/* runtime dependency.
Develop
npm ci
npm run build # build all workspaces
npm test # run all tests
Release
npm run build && npm test
npm -w apps/mcp publish
Version bumps happen in apps/mcp/package.json via a PR to this repository.
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.