entroly
Cut AI context cost without trusting the compressor. Every reduction is reversible, byte-exact recoverable, and carries an auditable receipt. Local-first, works
What is entroly?
Cut AI context cost without trusting the compressor. Every reduction is reversible, byte-exact recoverable, and carries an auditable receipt. Local-first, works through proxy, MCP, SDK, or agent wrapper.
Overview of entroly
What it is
Entroly is a local-first context control plane for AI coding tools. It selects evidence within a token budget, preserves recoverable omissions, and records selection decisions in receipts. Three properties: avoidable input cost (smaller requests can reduce input cost when routed through Entroly), recoverable originals (receipt-backed recovery retains source material locally), and auditable decisions (every decision comes with a receipt). It works without changing code on hosts with a verified prompt hook, and API traffic is intercepted only when routed through the Entroly proxy. The two commands entroly verify-claims and entroly simulate run locally with no API key required.
Key features
- Receipt-backed selection records: what was kept, what was omitted, and the handle that recovers the exact original bytes.
- Compression you can undo, on your own repository, in one command — without replacing your model or agent architecture.
- Local-first, works through proxy, MCP, SDK, or agent wrapper.
- Supports multiple platforms: Python (pip), Node (npm), Rust (source build), Homebrew, and Docker.
- Provides auditable receipts for every decision: what was kept, what was left out, and why.
- Agent continuity:
entroly-work resumereconstructs unfinished work from durable repository evidence, even if the previous agent left no handoff. - Compatible with VS Code, Claude Code, Cursor, Codex, OpenClaw, GitHub Copilot, Aider, and OpenAI/Anthropic-compatible apps.
Use cases
- Reducing input token costs when routing API traffic through the Entroly proxy.
- Recovering original source material from receipts when needed.
- Auditing what was kept and omitted in context selection for transparency.
- Switching coding agents mid-task without losing work, using
entroly-work resume. - Evaluating token-efficiency and context assurance on your own project with free local commands.
Getting started
bash pip install -U entroly cd /your/repo entroly verify-claims entroly simulate Both commands run locally without an API key. For a full quickstart with proxy and dashboard, run entroly go after installation.
License and pricing
Licensed under Apache-2.0. Open source and free to use.
Based on the project's public README and repository metadata.