Real Skill packageSource verifiedClawHub registry

model-intel-check(zjh)

黑盒检测某个 API 端点背后的模型是否"满血/智力正常"（中转站缩水鉴定）。当用户想验证某中转/代理/上游模型是否被换成弱模型、量化版或被剥离 thinking 时使用；也用于对比两个端点的同一模型。触发词如：满血、缩水、智力测试、鉴定模型、benchmark 这个端点。

Identity and source

Publisher attribution2641183145-ossregistry owner unverified by skillvetai
Functional categorySoftware Developmentautomatically inferred · 53% rule confidence
Package forminstruction with code13 recorded files
Canonical sourceClawHub registryclawhub:2641183145-oss:model-intel-check
Open canonical source ↗

Platform declarations

These states come from the source or distribution context. None of the entries below are SkillVetAI compatibility test results.

OpenClawnative officialProvenance: registry distribution

Independent structural checks

These checks parse the fixed package against dated platform rules. They do not execute the Skill or verify task behavior.

Claude Codepasses structure
Checker 0.1.0 · agent-skills-2026-08-13+claude-code-docs-2026-08-13 · 8/29/2026.claude/skills/model-intel-check

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

OpenAI Codexpasses structure
Checker 0.1.0 · agent-skills-2026-08-13+codex-docs-2026-08-13 · 8/29/2026.agents/skills/model-intel-check

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

OpenClawpasses structure
Checker 0.1.0 · agent-skills-2026-08-13+openclaw-docs-2026-08-13 · 8/29/2026skills/model-intel-check

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

Installation and inspection

This command is recorded from the source ecosystem and resolves the registry's latest release. The fixed release shown on this page should be inspected before adoption.

clawhub install @2641183145-oss/model-intel-check
clawhub inspect @2641183145-oss/model-intel-check --version 1.0.0

Security evidence

SkillVetAI static result: no findings detected

This automated, non-executing scan is bound to this release hash. It is not a safety certification and may contain false positives or false negatives.

Status
completed
Coverage
full text content
Files
13 / 13 inspected as text
Checked
8/29/2026, 4:46:57 PM
Scanner
0.1.3
Policy
1.0.3
5 inferred permission indicators
  • network access — automatically inferred
  • filesystem read — automatically inferred
  • filesystem write — automatically inferred
  • credential access — automatically inferred
  • external write action — automatically inferred
9 dependency and API indicators
  • api: clawhub.ai
  • api: deepmind.google
  • api: docs.bigmodel.cn
  • api: github.com
  • api: huggingface.co
  • api: openai.com
  • api: pypi.tuna.tsinghua.edu.cn
  • api: www.anthropic.com
  • api: x.ai
External clawhub result: suspicious

This is registry-supplied evidence for the recorded release, not an independent SkillVetAI scan. Check the canonical source for the full report, scanner versions, scope, and current moderation state.

Evidence checked
8/29/2026, 12:47:24 PM
Release binding
Matches this record
  • vt: clean
  • skillspector: suspicious
  • llm: suspicious

Recorded files

The catalog stores hashes and an inventory summary for change detection. It does not republish the package contents.

Package content hashsha256:cfca06d167491657b774b4297bb99b3732d21edbc443e7c29c53248c2c90d13b
Show up to 13 recorded paths
  • data/aime2025.jsonl
  • data/aime2026.jsonl
  • data/README.md
  • intel_check.py
  • make_data.py
  • references/anti-cheat.md
  • references/benchmarks.md
  • references/README.md
  • rerun_failed.py
  • serve.py
  • skill-card.md
  • SKILL.md
  • web/index.html

Source changelog

- Initial release of the "model-intel-check" skill for black-box evaluation of API endpoint models’ "full power" (no shrinkage or intelligence loss), using difficult public benchmarks (AIME 2025/2026 and GPQA Diamond). - Provides CLI scripts and a local web UI for running standardized benchmarking protocols against any OpenAI-compatible API endpoint. - Implements strict anti-cheat rules, standardized scoring protocol, and detailed verification/reporting flows, including official reference score lookup and extensive process documentation. - Includes procedures for handling API errors, transport failures, and manual review to ensure reliability and fairness of evaluation results. - Designed for users to identify endpoint model substitution, performance drops, or unexpected quantization, and to compare two endpoints’ underlying model quality.