Real Skill packageSource verifiedClawHub registry

Free Model Auditor(免费模型审计员)

审计 WorkBuddy 自定义模型注册表(models.json)中的免费模型:跨多个 OpenAI 兼容厂商新增可发现的免费模型、 剔除已转付费或失效的模型,保持注册表真实有效。当用户要求「审计自定义模型」「检查有没有新的免费模型」 「测试其余平台有无遗漏」「定期巡检模型清单」或希望对免费 API 模型做健康检查时使用。 本技能对海外平台执行 VPN 连通性门禁,按各厂商策略判定免费,活体实测每个候选,并自动把新增/移除差异应用到 models.json。

Identity and source

Publisher attributioniGenomedregistry owner unverified by skillvetai
Functional categorySoftware Developmentautomatically inferred · 53% rule confidence
Package forminstruction with code8 recorded files
Canonical sourceClawHub registryclawhub:igenomed:free-model-auditor
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 · 9/29/2026.claude/skills/free-model-auditor

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 · 9/29/2026.agents/skills/free-model-auditor

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 · 9/29/2026skills/free-model-auditor

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 @igenomed/free-model-auditor
clawhub inspect @igenomed/free-model-auditor --version 1.7.1

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
8 / 8 inspected as text
Checked
9/29/2026, 6:09:17 PM
Scanner
0.1.3
Policy
1.0.3
4 inferred permission indicators
  • network access — automatically inferred
  • filesystem read — automatically inferred
  • filesystem write — automatically inferred
  • credential access — automatically inferred
8 dependency and API indicators
  • api: api.agnes-ai.cn
  • api: api.siliconflow.cn
  • api: clawhub.ai
  • api: developer.amd.com.cn
  • api: generativelanguage.googleapis.com
  • api: integrate.api.nvidia.com
  • api: open.bigmodel.cn
  • api: token.sensenova.cn
External clawhub result: clean

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
9/29/2026, 6:50:05 AM
Release binding
Matches this record
  • vt: clean
  • skillspector: suspicious
  • llm: clean

Recorded files

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

Package content hashsha256:1d4ef51dbe97b1a062e141400306e93a7a1f656c3f38a52cba61c15c52159487
Show up to 8 recorded paths
  • manifest.yaml
  • references/apply_diff.py
  • references/platforms.md
  • references/resolve_paths.py
  • references/test_harness.py
  • skill-card.md
  • SKILL.md
  • templates/audit_report.md

Source changelog

# 免费模型审计员 (Free Model Auditor) · 更新说明 / Release Notes **版本 Version:** 1.7.1 **发布日期 Release Date:** 2026-09-24 **许可 License:** MIT-0 --- ## 一、核心更新 / Key Updates **中文** 1. **审计判据升级**:确立核心原则「免费 + 存活 ≠ 可用」。仅以单发请求返回 200 判定「免费可达」过松——免费公共端点常有极低的每秒请求数(rps)/ 每分钟令牌(TPM/RPM)上限,在 WorkBuddy 多请求对话模式下首句成功后即触发 429。新增候选须叠加「连续多请求不崩」可用性验证。 2. **剔除 `deepseek-flash`**:SenseNova 免费端点 `deepseek-flash` 实测为端点级限流(`429 EndpointTPMExceeded` / `rps exhausted`),在 WB 多轮/工具流中必现 429 不可用,已从可用集移除。 3. **工具增强**:`test_harness.py` 新增 `burst` 子命令(顺序发送 N 个请求,返回成功数/逐次状态/首个错误,`pass = 全部成功`),将可用性判据工具化。 4. **文档全中文化**:技能简介、SKILL.md、platforms.md、审计模板均为简明中文。 **English** 1. **Stricter audit criterion**: established the principle "free + alive ≠ usable". Judging "free & reachable" by a single 200 response is too loose — free public endpoints often have very low rps / TPM / RPM caps, so they hit 429 right after the first turn in WorkBud…

Release security diff

mediumCompared fixed releases 1.6.0 and 1.7.1; 0 finding and 0 permission indicators were added.

Both fixed releases were scanned under the current scanner and policy, so finding, permission and dependency changes are available.

Change reasons and limitations
  • dependency surface changed
  • content hash changed

Observed release history

These older immutable releases were observed by prior successful syncs. They remain recorded when a newer release becomes current.

1.6.08/30/2026sha256:77711c21b90a7b515c210249acf91355be5bf1e44bf10c0e237e7895acda1520
1.5.08/28/2026sha256:20c4c60be1a8876b1811a1fbc35772f162149198cf2ccb8327ca37e00cf25667