juyterman1000/entroly
The Context Engineering Engine. Your AI sees 5% of your codebase — Entroly shows it everything. 78% fewer tokens. Works with Cursor, Claude Code, Copilot, OpenClaw.
Entroly compresses entire codebases into context windows using variable-resolution indexing—critical files in full, supporting code as signatures, peripheral files as references—removing duplicates and boilerplate automatically. It operates as an MCP server for Cursor/Claude Code or a transparent HTTP proxy for other tools, with a Rust engine adding under 10ms overhead. The system learns from session outcomes via reinforcement learning to adjust context weights, includes 55 SAST security rules and codebase health analysis, and integrates deeply with OpenClaw for multi-agent budget allocation.
80 stars and 2,780 monthly downloads. Available on PyPI.
Stars
80
Forks
29
Language
Python
License
MIT
Category
Last pushed
Mar 27, 2026
Monthly downloads
2,780
Commits (30d)
0
Dependencies
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/juyterman1000/entroly"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.