ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Skill Vetter (by Azhua)

v1.0.0

Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope,...

⭐ 0· 1.6k·13 current·14 all-time
byRuei-Ci Wang@fatfingererr
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description match the content: SKILL.md is a human-readable vetting protocol and quick GitHub commands. No unrelated env vars, binaries, or installs are requested. _meta.json contains price and wallet info (expected for a catalog entry) but does not contradict the purpose.
ℹ
Instruction Scope
Instructions direct the agent/human to inspect all files in a skill repo, run harmless GitHub API curl commands, and look for clear red flags (network exfiltration, credential access, eval/exec, etc.). This is appropriate for vetting, but 'Read ALL files' requires that the vetting actor only access the skill's repository/workspace — do not conflate this with reading unrelated local/system files (e.g., ~/.ssh).
✓
Install Mechanism
No install spec and no code files — instruction-only. Lowest-risk install posture (nothing written to disk by the skill itself).
✓
Credentials
No required environment variables, credentials, or config paths are declared or used. The SKILL.md does not instruct using any secrets or service tokens.
✓
Persistence & Privilege
always is false and the skill is user-invocable. It does not request persistent presence or system modifications. Autonomous invocation is allowed by default on the platform, but this skill's instructions do not require or create elevated privileges.
Assessment
This skill is a manual vetting checklist and appears coherent and low-risk. Before using it: (1) confirm you're only inspecting the skill's repository/workspace (do not allow the agent to read unrelated local files like ~/.ssh or ~/.aws); (2) verify the author/repo yourself (stars, history, reviews) as recommended by the checklist; (3) treat the wallet/price in _meta.json as a catalog/payment detail, not a capability; (4) remember this is a guide for humans/agents — it cannot automatically prove a skill is safe. If you want automated checks, combine this checklist with sandboxed static analysis tools and human review for high-risk skills.

Like a lobster shell, security has layers — review code before you run it.

latestvk97ejcys1ajwhebys6ftg5h1fs81sf54

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments