create-claw-skill

Warn

Audited by Snyk on Mar 10, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 1.00). The included example skill "fetch-page" invokes the web_fetch tool at runtime to fetch an arbitrary URL provided as $1 (the user-supplied ), which injects remote page content into the model's context and therefore directly controls prompts—flagging the user-supplied URL passed to web_fetch (the /$1 argument).
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 03:40 PM