Back to skill
Skillv1.0.0
ClawScan security
Protocol Doc Auditor · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 22, 2026, 6:09 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requirements and instructions are coherent with its stated purpose (scanning docs for dangerous instructions); nothing requested is disproportionate or unexplained.
- Guidance
- This skill appears internally consistent. Before installing: (1) confirm you trust the skill publisher since source/homepage is unknown, (2) restrict the agent's network access if you run it against sensitive internal docs, and (3) remember the auditor flags patterns but cannot guarantee detection of novel or obfuscated attacks — for high-risk integrations follow up with manual review.
Review Dimensions
- Purpose & Capability
- okName and description claim a document-auditor for protocol docs; requiring curl (to fetch a URL) and python3 (to parse/analyze text) is reasonable and proportional to that purpose. No unrelated credentials, binaries, or config paths are requested.
- Instruction Scope
- okSKILL.md confines itself to analyzing provided URLs or text content for risky patterns (curl|bash, credentials in URLs, telemetry setup, etc.). It does not instruct the agent to read user files, environment secrets, or system configuration, nor to exfiltrate data to third parties.
- Install Mechanism
- okInstruction-only skill with no install spec or downloads. This is low-risk — nothing is written to disk by the skill itself.
- Credentials
- okNo environment variables, credentials, or config paths are required. The absence of sensitive requirements matches the auditor's function (text analysis of docs).
- Persistence & Privilege
- okalways is false and the skill does not request persistent system presence or modify other skills. Autonomous invocation is allowed by default but is not combined with other red flags here.