home / mcp / agenthotspot mcp server

AgentHotspot MCP Server

Provides a lightweight MCP server to search and access 6,000+ AgentHotspot MCP connectors for agent workflows.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "agenthotspot-agenthotspot-mcp": {
      "command": "python3",
      "args": [
        "-m",
        "agenthotspot_mcp"
      ]
    }
  }
}

AgentHotspot MCP Server enables your AI agents to search and discover OSS MCP connectors from the AgentHotspot marketplace. It provides a lightweight, MCP-compatible server you can run locally or integrate with clients to access thousands of connectors through simple commands or a standard MCP interface.

How to use

Run the MCP server locally and connect from your MCP client to search for connectors. You can start the server directly and then point your client to the local runtime for discovery and tool access. Use a standard MCP client workflow to query the AgentHotspot catalog with natural language and fetch tool descriptions or connector metadata.

How to install

Prerequisites and setup steps ensure you can run the server and integrate with your MCP clients.

# Prerequisites
- Python 3.10+
- An MCP-compatible client (Claude Desktop, OpenAI Agents SDK, custom agents, etc.)

# From Source
git clone https://github.com/AgentHotspot/agenthotspot-mcp.git
cd agenthotspot-mcp

# Install dependencies
pip install -r requirements.txt

# Install module
pip install -e .

Additional notes

Run the server independently to start a local MCP endpoint you can attach to from your agent framework. The server exposes an MCP interface suitable for standard MCP clients, and you can also integrate via common adapters like Claude Desktop or LangChain.

Available tools

search_connectors

Query the AgentHotspot catalog using natural language to discover connectors and capabilities.

analytics_dashboard

Overview of usage and performance metrics for MCP connector interactions.