Entroly
Local context-control plane for AI coding agents with token optimization, context receipts, and hallucination verification.
Entroly is a local context OS for AI coding agents that reduces input tokens by 70-95% on large repositories through intelligent ranking and selection of relevant evidence. It compresses supporting material while keeping originals recoverable, aligns with provider caches, and produces "Context Receipts" showing what was included or omitted. A verification gate checks AI answers against the provided evidence to catch hallucinations at near-zero cost.
Source
Repository: https://github.com/juyterman1000/entroly
Maintain Entroly?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[](https://getagentictools.com/mcp/juyterman1000-entroly?ref=badge) Related mcp
UniProf Universal CPU profiler for multiple languages and runtimes
Traditional Chinese Text Linting Enforces Taiwan Ministry of Education standards for Traditional Chinese text, catching Mainland Chinese regional drift in vocabula
Computer Use Control your computer with screenshots, mouse, and keyboard automation
PMB AI Local-first persistent memory for AI coding agents.
Strava Integrates with Strava's API to access and analyze fitness data, retrieve athlete profiles, activities, segments, routes, and club
Flyto Core Secure execution engine for AI agents with 300+ modules.