skill-vetter
Skill58 starsChanged 2 months agoMITView on GitHub
Security-first vetter for SKILL.md files — checks permissions, suspicious patterns, and metadata before installing or publishing skills. Useful for validating third-party skills for this repo.
26 other projects use a version of this.
Adopted by
Most recently taken up by
- CloudChef/atlasclaw-providerstheir own version, first seen Nov 8, 2026
- Soum-Code/master-skill-hubtheir own version, first seen Nov 8, 2026
- csfuwwc/md-skillstheir own version, first seen Nov 1, 2026
- weininghui/skillstheir own version, first seen Oct 30, 2026
- cadecode/agent-skillstheir own version, first seen Oct 20, 2026
- ai-workspace-lab/xworkspace-core-skillstheir own version, first seen Oct 17, 2026
Used with
Projects with it also use aminer-data-search, auto-target-tracker and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
dav-niu474Skillaminer-data-search
使用 AMiner 开放平台 API 进行学术数据查询与分析。当用户需要查询学者信息、论文详情、机构数据、期刊内容或专利信息时使用此 skill。 触发场景:提到 AMiner、学术数据查询、查论文/学者/机构/期刊/专利、学术问答搜索、引用分析、科研机构分析、学者画像、论文引用链、期刊投稿分析等。 支持 6 大组合工作流(学者全景分析、论文深度挖掘、机构研究力分析、期刊论文监控、学术智能问答、专利链分析)以及 28 个独立 API 的直接调用。 即使用户只说"帮我查一下 XXX 学者"或"找找关于 XXX 的论文",也应主动使用此 skill。
Also in 4 projects that use this
dav-niu474Skillauto-target-tracker
自动目标进度追踪器。在对话中检测到目标相关图片(笔记、进度、截图、记录)时,自动调用 VLM 识别关键信息并记录到目标日记。适用于学习管理、健身追踪、工作进度、习惯养成、创作记录等所有目标管理场景。
Also in 4 projects that use this
Liang-MHSkillstock_analysis
获取 A 股、基金和港股的实时行情和技术分析数据
Also in 4 projects that use this
- +9 more from dav-niu474
Other versions26 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 26
- freestylefly/wesight★ 969same name
- CloudChef/atlasclaw-providers★ 17same name
- choupiyang/ZAI_SKILLS★ 13same name
- dglijin-oss/chinese-metaphysics-skills★ 13same name
- ai-workspace-lab/xworkspace-core-skills★ 8same name
- b1mango/Agent-Skills★ 4same name
- csfuwwc/md-skills★ 4same name
- real-link-tech/reallink-skills★ 4same name
- seasky7c/legal-skills★ 4same name
- stvlynn/qclaw-skills★ 4same name
16 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository19 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- biomeSkill
- bunSkill
- changesetSkill
- clackSkill
- dry-refactoringSkill
- github-awesome-copilot-git-commitSkill
- hr-root-router-maintainingSkill
- hr-skills-maintainingSkill
- humanizerSkill
- jscpdSkill
- markdownSkill
- pdfSkill
- turboSkill
- typescriptSkill
- valibotSkill
- hr-skillsSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 26 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Check for dangerous shell commands or wide file write permissions
- Distinctive: line 18
Verify metadata.author and licensing sections
- Very specific: line 50
Run vetter as part of bun run validate or a pre-merge CI job
- Distinctive: line 51
Keep an allowlist of approved external hosts and commands
“Common” is measured: kept by at least 6 in 10 of the 26 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill Vetter
Supported tasks
- Validate SKILL.md frontmatter and required sections
- Check for dangerous shell commands or wide file write permissions
- Detect suspicious external URLs and credential leaks in content
- Detect hidden Unicode (zero-width characters, homoglyphs) used to hide injected instructions or spoof visible text
- Verify metadata.author and licensing sections
- Summarize permission scopes (read/write/commands) and risk level
- Produce remediation suggestions for failing checks
- Output a short pass/fail report for CI inclusion
- Suggest safe install steps or block listing
Key prompts
Validation
- "Validate this SKILL.md and report missing frontmatter fields: [content]."
- "Scan for bash commands that write to sensitive paths and list risky lines."
- "Check permission scope strings for overly broad write access: [permissions]."
- "Extract URLs from the skill and flag external hosts not on allowlist."
- "Scan this content for zero-width characters or homoglyphs that could hide instructions."
Remediation
- "Suggest fixes to reduce permission scope for this SKILL.md: [content]."
- "Rewrite dangerous command snippets into safe alternatives or sandboxed commands."
- "Produce a CI step snippet that fails on high-risk vetter issues."
- "Generate a short advisories list for reviewers before merging this skill."
Reporting
- "Produce a one-paragraph summary for maintainers explaining the vet result."
- "Create a Markdown checklist of vetter findings suitable for a PR comment."
- "Return a machine-readable JSON report with keys: passed, issues, severity."
- "Suggest whether to allow-install, require-review, or block-install based on findings."
Tips
- Run vetter as part of
bun run validateor a pre-merge CI job - Keep an allowlist of approved external hosts and commands
- Fail fast on write permissions targeting root-level or repo-critical paths
- Surface minimal, actionable fixes for authors to correct their SKILL.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

