seo-agentic

Pass

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill audits external websites by fetching and analyzing data like HTML, robots.txt, and llms.txt, which constitutes an ingestion surface for untrusted content.
  • Ingestion points: External URLs provided to the audit commands in SKILL.md.
  • Boundary markers: The instructions explicitly command the agent to treat fetched content as data, not as instructions, and to flag files that attempt to address the agent directly.
  • Capability inventory: The skill uses local scripts to perform analysis and writes configuration drafts to stdout for user review.
  • Sanitization: Implements a 'url_safety' utility to protect against SSRF and DNS-rebinding attacks.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the Lighthouse CLI tool via 'npx', which involves downloading a well-known package from the official npm registry at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 15, 2026, 01:46 PM