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.
# Add to your Claude Code skills
git clone https://github.com/juyterman1000/entrolyLast scanned: 5/23/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-23T06:34:21.167Z",
"semgrepRan": false,
"npmAuditRan": true,
"pipAuditRan": true
}See how entroly compares with popular alternatives.
entroly is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by juyterman1000. 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. It has 470 GitHub stars.
Yes. entroly passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/juyterman1000/entroly" and add it to your Claude Code skills directory (see the Installation section above).
entroly is primarily written in Python. It is open-source under juyterman1000 on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh entroly against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Historical maintainer-reported experiment, not a quality guarantee. The table below reports one 50K-token-budget run. Overlapping confidence intervals do not establish equivalence or rule out degradation. These results have not been independently reproduced here.
Model: gpt-4o-mini · Budget: 50K tokens · Wilson 95% CI · Reproduce: python -m bench.accuracy --benchmark all
| Benchmark | n | Baseline (95% CI) | Entroly (95% CI) | Retention | Benchmark Delta |
|---|---|---|---|---|---|
| NeedleInAHaystack | 20 | 100.0% [83.9–100%] | 100.0% [83.9–100%] | 100.0% | Baseline |
| GSM8K | 100 | 85.0% [76.7–90.7%] | 86.0% [77.9–91.5%] | 101.2% | +1.0% |
| SQuAD 2.0 | 100 | 84.0% [75.6–89.9%] | 83.0% [74.5–89.1%] | 98.8% | -1.0% |
| MMLU (4-way MCQ) | 100 | 82.0% [73.3–88.3%] | 85.0% [76.7–90.7%] | 103.7% | +3.0% |
| TruthfulQA (MC1) | 100 | 72.0% [62.5–79.9%] | 73.0% [63.6–80.7%] | 101.4% | +1.0% |
| LongBench (HotpotQA) | 100 | 57.0% [47.2–66.3%] | 59.8% [49.8–69.0%] | 104.9% | +2.8% |
The reported SQuAD score fell from 84% to 83%. Retention ratios above 100% can reflect sampling or model variability. The displayed intervals are historical reported values, not a validated paired comparison; Wilson intervals require binary outcomes and do not justify uncertainty for averaged partial-credit scores. Establishing non-inferiority needs a predefined tolerance, paired per-task outcomes, appropriate uncertainty estimates, and adequate sample size. These results cannot be extrapolated to more aggressive compression or other models.
19-fragment synthetic corpus · 300-token budget · 3 fixture queries · Reproduce: entroly benchmark
| Metric | RAW (Naive FIFO) | TOP-K (local baseline) | ENTROLY (Knapsack) |
|---|---|---|---|
| Avg fragments selected | 6.0 | 6.0 | 8.7 |
| Avg module coverage | 3.0 | 3.7 | 8.7 |
| Total SAST catches | 0 | 0 | 3 |
Entroly sees 8.7 modules where TOP-K sees 3.7 — it includes auth, payments, AND rate limiting. TOP-K misses the rate limiter. Full methodology, CIs, and reproduce commands →
Entroly includes the following research-oriented implementations. A module's presence does not establish production reliability, mathematical novelty, or independent validation; consult its implementation and evaluation limitations.
| Algorithm | What it does | Implementation |
|---|---|---|
| BIPT | Byte-level hallucination detection via Kolmogorov-inspired provenance tracing | provenance_tracer.py |
| NKBE | Nash-KKT multi-agent token budget equilibrium | nkbe.rs |
| Causal Context Graph | Intervention-aware fragment feedback learning | causal.rs |
| Cognitive Bus | ISA event routing with KL-divergence priority | cognitive_bus.rs |
| Resonance Matrix | Supermodular pairwise fragment value learning | resonance.rs |
| System 1 <> 2 | Dual-process verified-belief bridge (proxy <> vault) | coupling.py |
Use Entroly at the SDK, framework, proxy, MCP, plugin or agent boundary. A listed name is not automatically a claim that hosted subscription inference is intercepted; provider-bound savings exist only when the request traverses an Entroly-controlled route.
| Direct, tested paths | Guided or bounded paths |
|---|---|
| Vercel AI SDK middleware · OpenAI SDK · Anthropic SDK | Agno · Strands Agents · CrewAI · AutoGen |
| LangChain · LiteLLM · MCP | Claude Code on Vertex AI · Claude Code on Azure AI Foundry |
| OpenClaw · OpenCode | Claude Code in VS Code · VS Code Copilot · Grok |
Open the complete verified integration and operations hub →
AI coding assistants have a memory limit. Hand one your whole codebase and it gets slow, expensive, and distracted — like giving someone a 500-page manual when they only needed page 47.
Entroly finds page 47.
It sits between your code and the AI, reads everything, and passes along only the parts selected for the question. Three properties to evaluate on your task:
| 💰 Your bill goes down | Fewer words sent to the AI means a smaller invoice. How much depends on the job — see the [real numb |