Repowise is a codebase intelligence tool that provides code health scores, auto-generated documentation, git analytics, dead code detection, and architectural decision tracking through a Model Context Protocol interface.
The tool addresses the challenge of understanding and maintaining large codebases by combining multiple analysis approaches. It computes code health scores to quantify codebase quality, generates documentation automatically to keep it in sync with actual code, analyzes git history to surface patterns and trends, identifies unused code segments that accumulate technical debt, and captures architectural decisions in a queryable format. The Model Context Protocol integration allows both AI systems and human developers to access these insights programmatically, making the intelligence available to language models and custom tooling.
Repowise suits teams managing complex codebases who want visibility into code quality without manual auditing. It works well for projects where documentation tends to drift from implementation, where technical debt tracking is difficult to maintain, or where onboarding new developers requires understanding both current architecture and historical decisions. The combination of static analysis, git-based insights, and decision tracking makes it particularly valuable for teams using AI-assisted development, since the MCP interface feeds codebase context directly to language models.
The project shows active development with regular commits across multiple areas of the codebase. Work spans the core analysis engine, the MCP server implementation, documentation generation, and the web interface. The maintainers have established a public roadmap and respond to issues, indicating ongoing commitment to the tool's evolution. The project maintains a clear separation between analysis modules and integration points, suggesting a focus on extensibility and allowing different components to be used independently or combined.
