Repository Intelligence — Claude Code MCP server: analysis and setup
Analyzes code structure, dependencies, and security risks, with semantic search across a repository. Use it to explore unfamiliar code or inspect a project for vulnerabilities.
Open on GitHubWhat Repository Intelligence does
Repository Intelligence combines AST parsing, dependency graphs, and complexity metrics. It can look for hardcoded secrets, OWASP issues, and CVEs; semantic search helps locate code by meaning rather than exact keywords. It also provides reports focused on security, architecture, and technical findings.
The project includes Git history analysis and dead-code detection. It requires Python 3.9 or newer. The README describes single-machine, distributed, and cloud modes; its cloud option names AWS, GCP, and Azure. Cloud analysis therefore involves choosing and configuring infrastructure.
Repository Intelligence is listed for Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, and Zed. Project startup instructions are in QUICKSTART.md.
What it helps with
- find secrets and vulnerabilities in code
- map project dependencies and circular links
- inspect changes through Git history
- locate dead code and documentation gaps
- prepare a security or architecture report
Example prompts for the agent
- Scan this repository for secrets and OWASP vulnerabilities.
- Build a dependency graph and identify circular links.
- Find dead code and gaps in the documentation.
How to install Repository Intelligence
Install
The MCP registry has no launch command for it: see the repository README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about Repository Intelligence
How do I install Repository Intelligence in Claude Code?
The project's startup instructions are in QUICKSTART.md.
What does Repository Intelligence do?
It analyzes code and dependencies, searches for security issues and secrets, supports semantic search, and produces reports.
What does Repository Intelligence need to run?
The project requires Python 3.9 or newer. Its cloud mode lists AWS, GCP, and Azure.
