MCP Framework
A middleware system that connects large language models (LLMs) with various tool services through an OpenAI-compatible API, enabling enhanced AI assistant capabilities with features like file operations, web browsing, and database management.
What is MCP Framework?
MCP Framework is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to middleware system that connects large language models (llms) with various tool services through an openai-compatible api, enabling enhanced ai assistant capabilities with features like file operations...
A middleware system that connects large language models (LLMs) with various tool services through an OpenAI-compatible API, enabling enhanced AI assistant capabilities with features like file operations, web browsing, and database management.
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A middleware system that connects large language models (LLM
Use Cases
Maintainer
Works with
Installation
NPM
npx -y mcp-frameworkManual Installation
npx -y mcp-frameworkConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MCP Framework
MCP Framework is a Python middleware server that bridges large language models to multiple MCP tool services through an OpenAI-compatible /chat/completions HTTP API, enabling any LLM client that speaks the OpenAI format to gain access to tools like filesystem operations, web scraping, SQLite databases, and enhanced sequential reasoning. It supports multiple LLM backends simultaneously — OpenAI, Anthropic Claude, Azure OpenAI, Ollama, LM Studio, and Qwen — and manages multi-turn tool-calling sessions automatically, making it straightforward to extend any OpenAI-compatible chat application with powerful MCP-based capabilities.
Prerequisites
- Python 3.10 or later (or Docker for containerized deployment)
- API credentials for at least one supported LLM provider (OpenAI, Anthropic, Azure, Ollama, etc.)
- MCP server implementations for the tools you want to enable (time, filesystem, Firecrawl, SQLite, etc.)
- An MCP-compatible client or any application that calls the OpenAI chat completions API
Clone the repository
Download the MCP Framework source code from GitHub.
git clone https://github.com/Viking1726/mcp-framework.git
cd mcp-frameworkInstall dependencies
Install required Python packages from the requirements file.
pip install -r requirements.txtConfigure the framework
Edit config.json to set your LLM provider URL, default model, CORS settings, and the list of MCP servers to load. Each MCP server entry needs a command and optional arguments.
{
"server": { "host": "0.0.0.0", "port": 8000 },
"llm_service": {
"url": "https://api.openai.com/v1",
"default_model": "gpt-4o",
"timeout": 60
},
"mcp_servers": [
{ "name": "filesystem", "command": "npx", "args": ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"] },
{ "name": "sqlite", "command": "npx", "args": ["-y", "@modelcontextprotocol/server-sqlite"] }
],
"sessions": { "timeout": 3600, "max_sessions": 100 },
"logging": { "level": "INFO" }
}Start the server
Launch the FastAPI-based framework server directly with Python or via Docker Compose.
# Direct:
python run.py
# Docker:
docker compose up -dSend a chat request
Make an OpenAI-compatible chat completions request to the framework. It will automatically invoke MCP tools as needed and return the final response.
curl -X POST http://localhost:8000/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model": "gpt-4o", "messages": [{"role": "user", "content": "List files in /tmp"}]}'MCP Framework Examples
Client configuration
Use the MCP Framework as an OpenAI-compatible endpoint in any client that supports custom base URLs.
{
"mcpServers": {
"mcp-framework": {
"command": "npx",
"args": ["-y", "mcp-framework"],
"env": {
"OPENAI_API_KEY": "<your-llm-provider-key>"
}
}
}
}Prompts to try
The framework routes these requests to the configured LLM and automatically calls the appropriate MCP tools.
- "Read the contents of /tmp/notes.txt and summarize them"
- "Query the SQLite database at /data/app.db and show me the top 10 users by signup date"
- "Scrape the pricing page at https://example.com and extract all plan names and prices"
- "What time is it in Tokyo right now?"Troubleshooting MCP Framework
The server starts but tool calls fail with 'MCP server not found'
Verify that the command specified for each MCP server in config.json is installed and on your PATH. Run the command manually (e.g., 'npx -y @modelcontextprotocol/server-filesystem /tmp') to confirm it works before starting the framework.
LLM requests return 401 Unauthorized
The framework passes through your LLM provider's authentication. Ensure the API key or credentials for your LLM provider are set as environment variables (e.g., OPENAI_API_KEY) or configured in the llm_service section of config.json.
Sessions accumulate and memory usage grows over time
Set a reasonable 'timeout' value in the sessions config block (e.g., 3600 seconds) and limit 'max_sessions'. The framework uses these values to evict idle sessions automatically.
Frequently Asked Questions about MCP Framework
What is MCP Framework?
MCP Framework is a Model Context Protocol (MCP) server that middleware system that connects large language models (llms) with various tool services through an openai-compatible api, enabling enhanced ai assistant capabilities with features like file operations, web browsing, and database management. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MCP Framework?
Install via npm with the command: npx -y mcp-framework. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with MCP Framework?
MCP Framework works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MCP Framework free to use?
Yes, MCP Framework is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
MCP Framework Alternatives — Similar Developer Tools Servers
Looking for alternatives to MCP Framework? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up MCP Framework in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use MCP Framework?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.