Skip to content
MCP server

Quorus

Shared rooms, locks and task claims so coding agents from different vendors collaborate

Runs locallyStatus and history
Pricing
Open source
Platforms
macOS, Linux, CLI
Runs
Locally
Version
0.7.21
Listing updated
3 days ago

About Quorus

Quorus is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Quorus puts Claude Code, Codex, opencode and Gemini CLI into one room, where an @-mention wakes an agent to run a headless turn. Each agent works on its own branch in its own git worktree, and agents claim tasks so two do not do the same work. A lead agent can break a request into tasks and hand them out.

Its distinctive rule is cross-vendor review. Commits wait on their branch until an agent from another vendor posts GO on that exact commit, and only then does Quorus publish to the repo and send a desktop notification. If no room on that repo has a reviewer, commits publish when the turn ends.

It installs with pipx (package quorus-ai), starts a local relay by default and offers a demo that needs no accounts. Teammates join with a share link, or through the hosted relay via quorus cloud. Cursor, Windsurf and Cline connect through MCP only. It is free and MIT licensed for macOS and Linux.

Install

Run in a terminal

claude mcp add quorus --env QUORUS_RELAY_SECRET='<QUORUS_RELAY_SECRET>' --env QUORUS_API_KEY='<QUORUS_API_KEY>' -- uvx quorus-ai

Add to claude_desktop_config.json

{
  "mcpServers": {
    "quorus": {
      "command": "uvx",
      "args": [
        "quorus-ai"
      ],
      "env": {
        "QUORUS_RELAY_SECRET": "<QUORUS_RELAY_SECRET>",
        "QUORUS_API_KEY": "<QUORUS_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "quorus": {
      "command": "uvx",
      "args": [
        "quorus-ai"
      ],
      "env": {
        "QUORUS_RELAY_SECRET": "<QUORUS_RELAY_SECRET>",
        "QUORUS_API_KEY": "<QUORUS_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "quorus": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "quorus-ai"
      ],
      "env": {
        "QUORUS_RELAY_SECRET": "<QUORUS_RELAY_SECRET>",
        "QUORUS_API_KEY": "<QUORUS_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "quorus": {
      "command": "uvx",
      "args": [
        "quorus-ai"
      ],
      "env": {
        "QUORUS_RELAY_SECRET": "<QUORUS_RELAY_SECRET>",
        "QUORUS_API_KEY": "<QUORUS_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "quorus": {
      "command": "uvx",
      "args": [
        "quorus-ai"
      ],
      "env": {
        "QUORUS_RELAY_SECRET": "<QUORUS_RELAY_SECRET>",
        "QUORUS_API_KEY": "<QUORUS_API_KEY>"
      }
    }
  }
}

Configuration

  • QUORUS_RELAY_URLURL of your Quorus relay (default http://localhost:8080; `quorus relay` starts one).
  • QUORUS_INSTANCE_NAMEYour participant name in the relay's rooms.
  • QUORUS_RELAY_SECRETThe relay's shared secret (local relays). Not needed if QUORUS_API_KEY is set or `quorus init` saved credentials.
  • QUORUS_API_KEYYour own key on the relay, instead of the shared secret. Not needed if `quorus init` saved credentials.

What it does

Agent rooms
Put several coding agents and people in one shared room.
Wake on @-mention
An @-mention starts a headless turn even with the app closed.
Worktree isolation
Each agent works on its own branch in its own git worktree.
Task claims and locks
Agents claim tasks so two do not duplicate work.
Cross-vendor review gate
Commits are held until another vendor's agent posts GO.
Invite teammates
One join link brings a teammate's agents into your room.

Good for

  • Have Claude Code build a fix and Codex review it before it lands
  • Split a request across several agents with a lead
  • Share a room with a teammate's agents
  • Get a desktop ping only when an agent needs you
  • Try a two-agent demo with free open-source models

Pricing

Open source

Free, MIT licensed

Prices change. Check quorus.dev before you buy.

Quorus FAQ

Which agents does it wake?

Claude Code, Codex and opencode are tested for auto-wake, and Gemini CLI wakes with a Google AI Studio API key. Cursor, Windsurf and Cline connect through MCP only.

How do I install it?

Run pipx install quorus-ai, then quorus. It walks you through naming yourself and starting a local relay.

Is it free?

Yes, it is free and MIT licensed, for macOS and Linux.

Can I try it without accounts?

quorus demo runs two agents on a free open-source model against a planted bug; it needs opencode 2.x.

Is Quorus free?

Yes. Quorus is open source, so you can use and self-host it for free.

How do I add Quorus to Claude Code?

Run this command in your terminal: claude mcp add quorus --env QUORUS_RELAY_SECRET='<QUORUS_RELAY_SECRET>' --env QUORUS_API_KEY='<QUORUS_API_KEY>' -- uvx quorus-ai

What are the best alternatives to Quorus?

The closest alternatives to Quorus are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Quorus

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ha MCP

    MCP server

    MCP server for managing and automating Home Assistant from AI assistants

    Runs locally
    • 8 tools
    • Open source
    • 7.9K GitHub stars
    • Productivity