Skills/ClawHub/ai-eval-toolkit

ai-eval-toolkit@1.0.0

CAUTION· static only

Publisher claims: "LLM 评测工具链（可运行实现）——把评测方法论变成能直接跑的本地引擎：评测集管理（JSONL 建集/质量检查/规模统计）、幻觉检测引擎（数字一致性/引用校验/否定矛盾/关键论断互证四类规则检测）、RAG 指标计算（RAGAS 四指标的本地简化实现：忠实度/答案相关性/上下文精度/上下文召回）、回归对比（基线 vs 新结果差异判定）、报告生成与上线门禁（分场景得分/门禁判定/报告输出）。零依赖纯标准库，本地闭环不联网。与「LLM 质量评测」（方法论）互补——那个讲怎么做，这个给能跑的实现。面向 AI 工程师、测试与质量负责人。" Observed in sandbox: static analysis only — behavior not yet observed.

Registry ClawHubPublisher zhaoxinghua09-cell · publishing for 81 daysInstalls 427Last detonated 6d ago
What would fix thisWatchDispute verdict

Coverage

Static only; nothing was run yet.

The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.

Is [email protected] safe to install?

KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-08 and scored it 69/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Summary: static analysis only — behavior not yet observed.

Behavior trace

Queued for honeypot run · static analysis only
t+CategoryObserved action

This version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.

Static findings

MEDIUM
Python subprocess execution detectedverify/eval_toolkit_quality_test.py:23 · Python subprocess execution detectedr = subprocess.run([PY, os.path.join(TOOLS, "eval_toolkit.py")] + args,Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05

No other findings across 1,729 lines in 19 files.

Cost to the agent

Instructions loaded
≈1,042 tok
1 file · 19 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

11 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim zhaoxinghua09-cell: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Python subprocess execution detected≈ +2 on the totalPut hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim zhaoxinghua09-cellPropose a patch to the maintainer