RepoGuard

•
5
•

Enforce architectural guardrails, detect secret leaks, and generate AI context configurations for AI-assisted codebases.

RepoGuard acts as an automated architecture supervisor for AI-assisted development, preventing coding assistants from introducing code rot, bypassing architectural layers, or compromising type safety across TypeScript, Python, Go, and Rust. It automatically generates tailored context files such as .cursorrules, CLAUDE.md, and .windsurfrules, integrates directly into pre-commit hooks and CI/CD pipelines with SARIF reporting, and features a native Model Context Protocol (MCP) server that empowers AI agents to inspect architectural rules, perform audits, and review uncommitted diffs in real time.

Características Principales

01Real-time git diff auditing and actionable before-and-after refactoring suggestions
02Automated context file generation for Cursor, Claude Code, Windsurf, and GitHub Copilot
03Native Model Context Protocol (MCP) server providing audit tools, codebase health resources, and review prompts
0416 built-in guardrails checking layer boundaries, error handling, type safety, and secret leaks across TS, Python, Go, and Rust
05Pre-commit hook and CI/CD workflow integration supporting standard OASIS SARIF v2.1.0 output for GitHub Code Scanning
065 GitHub stars

Casos de Uso

01Auditing uncommitted git changes and PR diffs to prevent AI agents from breaking architectural boundaries
02Bootstrapping consistent, stack-tailored AI instruction rules across development teams
03Connecting AI IDE agents to a live MCP server for automated code health checks and refactoring plans