google-form

This MCP server implements tools to manage Google Forms via natural language through the CamelAIOrg Agent, exposing create_form, add_question, and get_form-related capabilities. The server translates MCP tool calls into Google Forms API requests, handling authentication and orchestration. You can use the web UI or the MCP API to create forms, add questions of various types (text, paragraph, multiple-choice, checkbox), and retrieve responses. The Agent interprets user requests, generates a sequence of MCP tool calls, and the server executes them against Google Forms, returning structured results back to the agent and ultimately to the UI.

To use it, first ensure the MCP server is running (via Docker in this setup). The web interface at the host port (default 5000) allows you to submit natural language requests like "Create a feedback form with 3 questions" and visualize the request flow. You can also directly POST to the MCP API endpoints to perform operations, and the Agent endpoint can be used to process raw natural language requests into tool calls when integrated with CamelAIOrg Agents.

Prerequisites:\n- Docker and Docker Compose installed on your machine.\n- Google Cloud Project with Google Forms API and Drive API enabled.\n- OAuth2 credentials (client ID, client secret) and a refresh token.\n- Optional: a domain/redirect URI configured for OAuth2 callback.\n\n1. Clone the repository:\nbash\ngit clone https://github.com/adarshp14-google-form-mcp-server.git\ncd adarshp14-google-form-mcp-server\n\n\n2. Prepare Google API credentials:\n- Create a Google Cloud project, enable Forms and Drive APIs, and create OAuth2 credentials.\n- Obtain a refresh token using the OAuth 2.0 Playground and place values in a .env file (see step 3).\n\n3. Create a .env file with required environment variables:\nbash\n# Google API Credentials\nGOOGLE_CLIENT_ID=your_client_id_here\nGOOGLE_CLIENT_SECRET=your_client_secret_here\nGOOGLE_REFRESH_TOKEN=your_refresh_token_here\n\n# Server Configuration\nFLASK_ENV=development\nPORT=5000\nDEBUG=True\n\n# CamelAIOrg Agents Configuration\nAGENT_ENDPOINT=http://agents:5001/process\nAGENT_API_KEY=your_agent_api_key_here\n\n\n4. Build and run with Docker Compose:\nbash\ndocker-compose up --build\n\nThis will start the MCP Server (port 5000) and CamelAIOrg Agents (port 5001).\n\n5. Verify the setup:\n- Access the web UI at http://localhost:5000\n- Use the MCP endpoints (e.g., POST http://localhost:5000/api/process) to issue tool calls.\n

Tips and considerations:\n- Ensure Google OAuth2 credentials and refresh token are kept secure. Keep credentials in a .env file or use a secret manager in production.\n- The MCP schema exposes tools such as create_form, add_question, and get_form. When integrating with CamelAIOrg, the agent will plan a sequence of these tools from a natural language request.\n- If you modify the Google Forms API permissions or scopes, re-authenticate and refresh tokens accordingly.\n- Docker Compose is used for easy orchestration; ensure the docker-compose.yml in the repo matches your deployment environment.\n- For debugging, check the MCP server logs and the agent service logs to trace request flow and handle errors gracefully.\n