Awesome AI Agents › AI Agent Gateway

golf-mcp/golf

⭐ 840 Python added to this list on 2025-06-16 repository created 2025-02-24

Golf is a production-ready framework designed to simplify the development, deployment, and scaling of MCP (Modular Control Program) server applications that power AI agents. It provides a structured and efficient way for developers to build secure AI agent infrastructure by defining server capabilities such as tools, prompts, and resources as Python files organized within a conventional directory structure. Golf automatically discovers, parses, and compiles these components into a runnable FastMCP server, significantly reducing boilerplate code and accelerating the development process. The framework is tailored for developers who want to focus on implementing the logic of their AI agents without dealing with the complexities of server setup and integration. It supports features essential for production environments, including authentication, observability, debugging, telemetry, and runtime management. Golf's configuration is highly customizable through a central `golf.json` file, allowing users to specify server name, port, transport protocols (such as Server-Sent Events, streamable HTTP, or standard I/O), telemetry options, and health check endpoints. Golf projects follow a clear directory structure with dedicated folders for tools, resources, and prompts, each containing Python files that define individual components. The framework uses the module docstrings for component descriptions and infers input parameters and output schemas from function signatures and Pydantic models, enabling automatic registration and integration of components. It also supports nested directories and shared code modules to organize complex projects efficiently. The project includes a command-line interface (CLI) for initializing new projects, building, and running the development server, making it easy to get started and iterate quickly. Golf is designed to be extensible and scalable, suitable for real-world applications that require robust AI agent infrastructure. Comprehensive documentation and example code are provided to assist developers in leveraging the framework effectively. In summary, Golf is an all-in-one MCP server framework that streamlines the creation and management of AI agent servers, offering built-in support for security, observability, and telemetry, while enabling rapid development through automatic component discovery and minimal boilerplate.

https://github.com/golf-mcp/golf

agent-runtimeaiai-agentai-agent-infrastructureai-agent-toolsai-agentsai-platformaiagentsauthauthenticationauthorizationclicomponent-discoveryconfigurationdebuggingextensiblefastmcpframeworkhealth-checkhttp-streamingintegrationmcpmcp-frameworkmcp-servermcp-serversmcp-toolsmcpsmodularobservabilityproduction-readypromptspydanticpythonresourcesruntimescalablesecureserver-sent-eventsserver-setupstandard-i/otelemetrytools