The goal is to build a codeforces's mcp server for users to get codeforces info by LLM.
目标是搭建一个codeforces的mcp服务器，去帮助用户仅使用大语言模型就可以获取codeforces的信息
本地测试 mcp-server 用
mcp dev server.pyMCP.Pizza Chef: BaseAI998
Codeforces-MCP is an MCP server designed to expose Codeforces platform data to large language models. It enables users to query contest information, user profiles, and problem details from Codeforces directly through LLMs, facilitating real-time, structured access to competitive programming data without leaving the model interface.
Use This MCP server To
Query Codeforces contest standings and results via LLM Retrieve user profile and rating history from Codeforces Fetch problem statements and details for competitive programming Integrate Codeforces data into AI-assisted coding workflows Enable LLMs to answer Codeforces-related questions in real time
README
Codeforces-MCP FAQ
How do I start the Codeforces-MCP server locally?
Run `mcp dev server.py` in your terminal to start the server locally for testing.
Can this MCP server provide real-time contest updates?
Yes, it fetches live data from Codeforces to provide up-to-date contest information.
Is authentication required to access Codeforces data through this MCP server?
Typically, public Codeforces data is accessible without authentication, but some endpoints may require user credentials.
How does this server handle large data requests from Codeforces?
It structures and paginates data to efficiently serve LLM queries without overwhelming the model.
Can I extend this MCP server to support other competitive programming platforms?
Yes, the architecture allows adding adapters for other platforms similar to Codeforces.
Which LLM providers are compatible with Codeforces-MCP?
It works with OpenAI, Anthropic Claude, and Google Gemini models via the MCP protocol.