Repowise
Codebase intelligence for AI-assisted engineering teams: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
What is Repowise?
Repowise is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to codebase intelligence for ai-assisted engineering teams: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via mcp.
Codebase intelligence for AI-assisted engineering teams: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Codebase intelligence for AI-assisted engineering teams: cod
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx repowiseConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Repowise
Repowise is a codebase intelligence MCP server that pre-builds five analysis layers—dependency graphs, git hotspot analytics, auto-generated per-module documentation, architectural decision records, and 25-metric code health scores—and exposes them through nine MCP tools. AI coding agents like Claude Code connect to Repowise to answer deep codebase questions without repeatedly scanning files, reducing token consumption by up to 96% and file reads by 89%. Teams use it for dead code detection, defect risk assessment, and onboarding new engineers or AI agents to large, unfamiliar codebases.
Prerequisites
- Python 3.11+ installed (pip or uv package manager)
- Git repository (single repo or a workspace of multiple repos)
- An Anthropic or OpenAI API key for LLM-powered documentation generation (or Ollama for fully offline use)
- An MCP-compatible client such as Claude Code or any MCP-enabled IDE extension
Install Repowise
Install the repowise package from PyPI using pip or the faster uv tool installer.
pip install repowise
# or with uv:
uv tool install repowiseInitialize the intelligence layers
Run repowise init from inside your project directory. This one-time command builds all five intelligence layers: dependency graph, git analytics, per-module docs, architectural decisions, and code health scores. Use --index-only to skip LLM calls.
cd your-project
repowise init
# For a workspace of multiple repos:
repowise init .Start the MCP server
Launch the Repowise MCP server and local dashboard. The server listens on stdio by default for MCP clients, and opens a web dashboard for direct inspection.
repowise serveConfigure your MCP client
Add Repowise to your MCP client configuration, pointing to the project directory as an argument so it serves the correct codebase.
Keep the index fresh with incremental updates
After committing new code, run repowise update for a fast incremental refresh (typically under 30 seconds), so all nine MCP tools reflect the latest changes.
repowise updateRun standalone health and risk checks
Use the CLI directly for CI/CD integration. The health command shows per-file scores, risk assesses a branch or PR for defect likelihood, and dead-code detects unreachable functions.
repowise health
repowise risk main..feature-branch
repowise dead-codeRepowise Examples
Client configuration (Claude Code)
Add Repowise to Claude Code's MCP server list, pointing to the indexed project directory.
{
"mcpServers": {
"repowise": {
"command": "repowise",
"args": ["serve", "/path/to/your/project"]
}
}
}Prompts to try
With Repowise connected, ask your AI agent deep architectural and quality questions about the codebase.
- "What are the highest-risk files in the current branch compared to main?"
- "Show me the dead code in the payments module"
- "What architectural decisions were made about the database layer?"
- "Which files have the lowest code health scores and what are the main issues?"
- "Summarize the documentation for the auth module"Troubleshooting Repowise
repowise init fails with an API key error
Set your LLM provider key as an environment variable before running init: export ANTHROPIC_API_KEY=sk-... or export OPENAI_API_KEY=sk-.... Alternatively, configure Ollama in .repowise/config.yaml and use --index-only to skip LLM doc generation.
The MCP tools return stale results after recent commits
Run repowise update from the project directory to trigger an incremental refresh of the intelligence layers. For fully automated freshness, add repowise update to your post-commit git hook.
Multi-repo workspace init is slow or incomplete
Run repowise init from the parent workspace directory (repowise init .) so it discovers all nested git repos. Use --index-only to skip LLM calls for faster indexing when documentation generation is not needed immediately.
Frequently Asked Questions about Repowise
What is Repowise?
Repowise is a Model Context Protocol (MCP) server that codebase intelligence for ai-assisted engineering teams: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via mcp. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Repowise?
Follow the installation instructions on the Repowise GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Repowise?
Repowise works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Repowise free to use?
Yes, Repowise is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
Repowise Alternatives — Similar Developer Tools Servers
Looking for alternatives to Repowise? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Repowise in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Repowise?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.