hiveflow

The HiveFlow MCP server acts as a bridge between your AI assistants (such as Claude or Cursor) and the HiveFlow automation platform. Once configured, it exposes a set of Flow Management and MCP Server Management tools that your AI assistant can call to create, list, and execute automation flows, as well as manage the MCP server configuration itself. The server communicates with HiveFlow via the API key you provide, enabling secure access to your flows, executions, and server settings. Use the provided tool commands to manage flows (create_flow, list_flows, get_flow, execute_flow, pause_flow, resume_flow, get_flow_executions) and to manage MCP servers (list_mcp_servers, create_mcp_server), establishing a seamless automation workflow between natural language prompts and your automation platform.

Prerequisites:

• Node.js and npm installed on your machine
• Internet access to install the MCP server package

Step 1: Install the MCP server globally

npm install -g @hiveflow/mcp-server

Step 2: Configure your MCP client with the HiveFlow server Create or update your MCP configuration (e.g., .cursor/mcp.json) with the following structure:

{
  "mcpServers": {
    "hiveflow": {
      "command": "npx",
      "args": ["-y", "@hiveflow/mcp-server"],
      "env": {
        "HIVEFLOW_API_KEY": "your-api-key-here",
        "HIVEFLOW_API_URL": "https://api.hiveflow.ai"
      }
    }
  }
}

Step 3: Start using the MCP server

• If you installed globally, your client will resolve the hiveflow MCP server as configured above.
• Ensure the API key and URL are correct and the HiveFlow instance is reachable.

Optional local development setup (alternative same config):

{
  "mcpServers": {
    "hiveflow": {
      "command": "npx",
      "args": ["-y", "@hiveflow/mcp-server"],
      "env": {
        "HIVEFLOW_API_KEY": "your-api-key-here",
        "HIVEFLOW_API_URL": "http://localhost:5000"
      }
    }
  }
}

Tips and common issues:

• Ensure HIVEFLOW_API_KEY is valid and has the necessary permissions for your flows.
• If you encounter "HIVEFLOW_API_KEY is required", verify that the key is set in the MCP client configuration and is not overwritten by environment variables.
• The default API URL is https://api.hiveflow.ai; override it if you are using a self-hosted or staging instance by setting HIVEFLOW_API_URL.
• For debugging, enable logging by setting DEBUG=hiveflow-mcp:* in your environment.
• If the MCP server cannot be found, verify the npm global install and that your MCP client references the correct package name and version.
• This MCP server design assumes a standard Node.js/npm environment and uses npx to fetch the server package on demand.