Side-by-side comparison of two Model Context Protocol servers — install paths, tools, quality signals, and directory engagement so you can pick the right one for Claude, Cursor, and other MCP clients.
Read and analyze GitHub repositories with your LLM
Selects the minimum code context an LLM needs to understand a git diff: graph-based fragment selection under a token budget, deterministic, 30+ languages. Published on PyPI and in the MCP Registry.
Quality signal
36/100 (Emerging)
41/100 (Fair)
Install path
Remote · high
uvx · high
Engagement
4 0 0 313
1 0 0 2
Tools
Clones GitHub repos with caching via hash-based temp directoriesGenerates Unicode tree representation of directory structureReads multiple specified files with error handlingCleans up temporary clones after processingExposes tools via MCP server interface