ACI.dev is an open-source AI agent tool infrastructure that provides multi-tenant authentication, permission management, and dynamic discovery functions for over 600 pre-integrated tools. It supports calls through a unified MCP Server or a lightweight SDK, simplifying the development of production-level AI agents.

This is a server that implements the MCP code execution mode. Through a single-tool bridge and zero-context discovery mechanism, it reduces the overhead of MCP tool calls from tens of thousands of tokens to about 200 tokens and securely executes Python code in the root container, supporting data science and secure isolation.

This project provides support for Model Context Protocol (MCP) tool calls for LangChain, enabling interaction with AI models through the MCP toolkit, such as file reading and summary generation.

Pipelock is an open-source AI agent firewall that provides security protection with zero runtime dependencies through a single binary file. It adopts an architecture with separated capabilities and deploys a nine-layer scanning pipeline between the agent and the Internet to prevent secret leakage, prompt injection, and malicious tool calls. It supports multiple proxy modes (Fetch proxy, forward proxy, WebSocket proxy) and MCP proxy, and provides functions such as an emergency stop switch, audit logs, and SIEM integration.

This project builds a bridge between Claude Code and Google Gemini AI, enabling direct calls to Gemini in the Claude Code environment for Q&A, code review, and creative brainstorming, providing a convenient AI collaboration tool.

An MCP server for batch - reading memory bank files, which reduces the frequency of AI tool calls and saves time and costs by loading all files in the directory at once.

This project is a bridging tool that connects Claude Desktop to remote servers. It uses Node.js to convert protocols from stdio to HTTP SSE, supporting service discovery and tool calls of the MCP protocol.

The Linkup MCP Server is a tool used to integrate Linkup's web search functionality with AI models supporting function calls through the MCP protocol.

The Infisical Model Context Protocol (MCP) is a service protocol that integrates with the Infisical API through function calls. It supports multiple tool operations such as creating, deleting, and updating keys, and provides a debugging tool called MCP Inspector.

This project provides a FastMCP server for batch execution of file and folder operations, including reading, creating, moving, deleting, etc., supporting centralized exception handling and batch tool calls.

A reference project implementing the Model Context Protocol (MCP), providing a standard for tool calls between LLMs and applications, including local and remote call implementations, and integrating AWS Bedrock and Claude 3.7.

DiffuGen is an advanced local image generation tool that integrates the MCP protocol, supports multiple AI models (including the Flux and Stable Diffusion series), and can directly generate high-quality images in the development environment. It provides flexible configuration options, multi-GPU support, and can be integrated with multiple IDEs through the MCP protocol. It also provides an OpenAPI interface for external calls.

A local multi - expert intelligent agent scheduling system that supports automatic generation of experts, tool calls, and knowledge base expansion, used for a graduation project.

The MCP client project is used for advanced interaction with AI models and services, supporting tool calls and server management.

The Flutter MCP Server is an open - source implementation of the Model Context Protocol (MCP) for the Dart/Flutter ecosystem. It provides a unified API interface for AI assistants and developer tools, supports remote calls to the Dart/Flutter development toolchain, and enables secure and programmable automated development processes.

This project provides code analysis and editing capabilities through the LSP protocol, supports MCP tool calls for TypeScript/JavaScript, includes functions such as file operations, symbol lookup, and error checking, and has security features such as path restrictions.

OneTool is a unified MCP server that exposes over 100 tools through the Python API, significantly reducing token consumption and cost when AI calls tools, avoiding context pollution, and supporting multiple functions such as web search, database, file operations, and chart drawing.

The Feishu MCP Server is a service that implements the Model Context Protocol, supporting interaction with the Feishu API through tool calls and mainly providing Bitable operation functions.

Bookworm is a toolset for processing docs.rs documentation, including functions such as downloading, indexing, and querying documentation, and provides an MCP service interface for LLM calls.

The Gridly MCP Server is a server-side tool for managing Gridly API projects, providing functions such as project management, grid operations, and database management, and supporting calls through MCP clients such as Claude Desktop.

The MetaTool MCP Server is a proxy server used to connect multiple MCP servers and forward tool calls to the appropriate servers. It is used in conjunction with the metatool-app to provide GUI tool management functions. The project is deprecated, and the Node.js alternative version is recommended.

The Vapi MCP Server is a tool that integrates the Vapi API through function calls, supporting voice assistant calls and appointment management through Claude Desktop.

The WeChat Mini Program Automated Testing MCP Server realizes automated control of the WeChat Developer Tools through FastMCP and miniprogram - automator, providing a toolset for page navigation, element operation, API calls, etc.

An MCP protocol test server implemented in Rust, supporting basic commands and tool calls.

The Kagi MCP Server is a tool integration service that provides search and summarization functions for Claude Desktop, supporting API calls to Kagi's search and summarization engines.

This is a Clojure SDK for creating a Model Context Protocol server, providing core components, communication processes, and example implementations, supporting tool calls, resource reading, and prompt retrieval.

A hybrid retrieval and navigation tool for RimWorld source code and XML definitions, combining lexical, semantic, and graph structures, supporting AI assistant calls

The Model Context Protocol (MCP) is an open - source protocol released by Anthropic, aiming to solve the standardization problem of connecting large language models (LLMs) with external data sources and systems. It provides a structured framework that enables models to integrate and utilize external context, thereby expanding their capabilities and improving response accuracy. MCP supports functions such as knowledge expansion, external tool calls, and pre - written prompts.

An MCP hub server for connecting to and managing other MCP servers, breaking through the tool quantity limit and optimizing tool calls

The MCP test client is a tool for testing Model Context Protocol servers, providing simple interfaces for tool calls and response validation.

The Infisical Model Context Protocol (MCP) Server provides a set of tools for integrating with the Infisical API through function calls, supporting operations such as creating, deleting, updating, and querying keys. It is suitable for key management and project collaboration.

This project demonstrates how to deploy a remote MCP server on Cloudflare Workers and implement the OAuth login function. It supports local development and debugging, connection testing through the MCP Inspector, and integrating the Claude desktop client into the MCP service for math tool calls. Finally, it provides a complete process for deploying to Cloudflare and remote connection.

An MCP server that provides Polygon PoS blockchain interaction tools for Claude AI, supporting contract calls, token transfers and other functions.

MCP BatchIt is a batching server for MCP tool calls. By aggregating multiple MCP tool calls into one request, it significantly reduces the communication overhead and token consumption of AI agents.

The MCP Gateway is a reverse proxy server used to centrally manage and forward client requests to multiple MCP servers, supporting SSE streams and tool calls.

A minimalist client tool for testing tool calls in the MCP server

The MCP PostgreSQL Server is a standardized interface service that allows AI models to interact with PostgreSQL databases through tool calls, supporting functions such as connection management, query execution, and table structure viewing.

The MCP server for ROS2 enables AI tools to connect to ROS2 nodes, topics, and services through the standard MCP protocol. It provides functions such as topic subscription and publishing, service calls, and action control, supports nested fields and automatic type discovery, and simplifies the integrated development of AI and ROS2.

This project is an MCP server integrated with the Stable Diffusion image generation function, providing image generation services for AI agents. It supports generating images through the MCP protocol or direct API calls and includes development and debugging tools.

NCP is an intelligent orchestrator for the MCP protocol that manages multiple MCP servers in a unified manner. It automatically routes tool calls through semantic understanding, significantly reducing the cognitive burden and token consumption of AI tools, and improving the work efficiency and response speed of AI assistants.