evalops/deep-code-reasoning-mcp

View on GitHub ↗

A Model Context Protocol (MCP) server that provides advanced code analysis and reasoning capabilities powered by Google's Gemini AI

107 ★13 forksTypeScriptUpdated 24d ago

What you need to know

An MCP server pairing Claude Code with Google Gemini AI for complementary code analysis, including AI-to-AI conversations and 1M-token context debugging.

Install

git clone, npm install, set GEMINI_API_KEY in .env, npm run build

Usage

  • Register as a stdio MCP server; Claude escalates complex analysis (long traces, cross-service failures) to Gemini, which responds within its 1M-token context.

Key features

  • Conversational AI-to-AI analysis tools
  • Execution flow tracing and cross-system impact analysis
  • Performance modeling and hypothesis testing
  • Long context support for large codebases

Best for

Developers debugging distributed systems with a multi-model Claude plus Gemini workflow.

Caveats

  • Requires a Gemini API key; code and logs are sent to Google's Gemini API.
Clients: Claude Desktop · Cursor

Reviewed 2026-08-11

Topics

aiai-toolsclaudecode-analysiscode-intelligencecode-reasoningdebuggingdeveloper-toolsdistributed-systemsgeminihypothesis-testingllmmcpmcp-servermodel-context-protocolmulti-modelnodejsperformance-analysissemantic-analysistypescript
Stars
107★
Forks
13
Language
TypeScript
License
MIT
Created
2025-06-11
Last push
2026-08-11