pseo-audit
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill specifies a clear interface for SEO analysis using a dedicated CLI tool or MCP interface. It focuses on auditing programmatic SEO templates by analyzing search performance data and technical samples (crawls and URL inspections) according to a predefined machine contract. No evidence of prompt injection, data exfiltration, or unauthorized command execution was found.\n- [SAFE]: The skill processes external data from Google Search Console, sitemaps, and web page crawls. While this introduces a surface for indirect prompt injection, the instructions provide a structured contract and specific interpretation rules to guide the agent's behavior.\n
- Ingestion points: Data entering the agent via
seo_run_report(GSC data), sitemap XML files, and text content from sampled page crawls.\n - Boundary markers: The skill instructs the agent to read
structuredContentas the machine contract and provides a strict rubric for interpreting health verdicts.\n - Capability inventory: The associated
seotool has the capability to read search analytics, crawl external URLs, and export results to the local filesystem (./seo-exports/pseo).\n - Sanitization: No explicit sanitization of external content is mentioned in the skill instructions, as it relies on the underlying tool's implementation.
Audit Metadata