seo-drift
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes untrusted HTML content, meta tags, and JSON-LD schema from external URLs which could contain malicious instructions designed to influence the agent's behavior during comparison analysis.\n
- Ingestion points: External page content is fetched via
scripts/fetch_page.pyand parsed byscripts/parse_html.py.\n - Boundary markers: The instructions lack explicit delimitation or instructions for the agent to ignore potential directives embedded within the fetched SEO elements.\n
- Capability inventory: The skill can execute local Python scripts, perform SQLite database operations, and write report files to the filesystem.\n
- Sanitization: While SSRF protection is mentioned via
validate_url(), there is no explicit mention of sanitizing or escaping the content parsed from the target URLs before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill instructions require the agent to execute several local Python scripts (e.g.,drift_baseline.py,drift_compare.py,drift_report.py) to perform data capture, analysis, and report generation.
Audit Metadata