Agent Hub

Provides a universal coordination layer for MCP-compatible AI agents to collaborate across projects and share context.

typescript

24

GitHub Stars

typescript

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "gilbarbara-agent-hub-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "agent-hub-mcp@latest"
      ],
      "env": {
        "AGENT_HUB_DATA_DIR": "YOUR_DATA_DIR"
      }
    }
  }
}

Agent Hub MCP is a universal coordination layer that lets any MCP-compatible AI agent collaborate across projects, share knowledge, and coordinate tasks. It removes silos between agents and enables smooth multi-agent workflows across your AI ecosystems.

How to use

You connect your MCP-compatible AI assistants to Agent Hub MCP so they can coordinate on features, tasks, and context. Once connected, agents can register themselves, sync workloads, and share results across frontend and backend or across different platforms. The hub maintains feature boundaries and preserves context for each collaboration so you can work end-to-end without losing state.

How to install

Prerequisites you need before installation:

Node.js 22+ installed on your system.
An MCP-compatible AI assistant (for example Claude Code, Qwen, Gemini, Codex) ready to connect to an MCP network.

Follow these steps to set up the Agent Hub MCP server locally and prepare your assistants to join the hub.

Step 1 Add Agent Hub MCP to your AI assistant

{
  "mcpServers": {
    "agent-hub": {
      "command": "npx",
      "args": ["-y", "agent-hub-mcp@latest"]
    }
  }
}

Step 2 Install custom commands (recommended)

git clone https://github.com/gilbarbara/agent-hub-mcp.git /tmp/agent-hub-mcp
mkdir -p ~/.claude/commands/hub
cp /tmp/agent-hub-mcp/commands/markdown/*.md ~/.claude/commands/hub/

git clone https://github.com/gilbarbara/agent-hub-mcp.git /tmp/agent-hub-mcp
mkdir -p ~/.qwen/commands/hub  # or ~/.gemini/commands/hub
cp /tmp/agent-hub-mcp/commands/toml/*.toml ~/.qwen/commands/hub/

Step 3 Restart your AI assistant

Close and reopen your AI assistant completely for changes to take effect.

Step 4 Verify installation

Register your agent and verify the hub connection. You can use the hub commands in your assistant to register and check status.

If you do not see a successful registration, confirm the MCP server configuration is reachable and restart your AI assistant.

Automatic Message Notifications (Optional)

You can enable automatic notifications when other agents send messages by adding a hook to your assistant’s settings. After setup, you will see unread message counts and be prompted to sync.

Available tools

register_agent

Register or reconnect an agent to the hub and establish a persistent identity across restarts.

send_message

Send a message to other agents or fetch messages from the hub for collaboration.

sync

Synchronize agent workloads and check for updates across features and tasks.

get_hub_status

Get an overview of hub activity and current collaboration state.

create_feature

Create a multi-agent feature to coordinate work across agents and repositories.

create_task

Create and delegate tasks within a feature to specific domain experts.

update_subtask

Update the status and output of a subtask as work progresses.

get_feature

Retrieve detailed information about a feature, including subtasks and outputs.