AI BriefingRepowise, an Open-Source MCP Tool That Boosts Claude Code Efficiency
Key point
Repowise, an MCP tool that improves Claude Code's task accuracy and cuts costs through code health scores, has been released.
Details
Repowise is an open-source tool that calculates a file's Code Health Score based on the AST (Abstract Syntax Tree) and Git history.
This tool does not use an LLM directly, instead leveraging 25 deterministic biomarkers such as complexity, duplication, developer churn, and untested hotspots. Through this, it predicts with high accuracy which files are likely to produce bugs.
In particular, it operates as an MCP (Model Context Protocol) layer for Claude Code, allowing the AI agent to recognize a file's risk level before modifying it. This helps prevent the model from making mistakes by touching fragile files and encourages it to identify dependencies first.
Actual performance tests confirmed the following improvements compared to standard Claude Code:
- 49% reduction in tool calls
- 89% reduction in file reads
- 36% reduction in operating costs
It is currently drawing attention with over 2.2K stars on GitHub and is provided under the AGPL-3.0 license.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Atlas: Open-Source Source Control Tool for Multi-Agent Coding WorkflowsWhiteboard Launches Open-Source VS Code Fork for Human-AI Collaborative Software Design
Token-Space Font Compiler Generates Fonts with Fixed Token Widths