search-page-audit
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONNO_CODE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches HTML source, rendered content, and configuration files (robots.txt, sitemap.xml, llms.txt) from user-provided external URLs to conduct audits.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by processing external web content.\n
- Ingestion points: External HTML source, rendered content, and site configuration files (robots.txt, sitemap.xml, llms.txt) entering the agent context via instructions in SKILL.md.\n
- Boundary markers: Absent; the skill lacks delimiters or instructions to ignore embedded commands within the fetched data.\n
- Capability inventory: The skill uses network-read capabilities (fetching URLs) to aggregate data for analysis.\n
- Sanitization: Absent; no sanitization or escaping of the retrieved content is specified.\n- [NO_CODE]: The skill consists of instructional markdown without any executable code or binary components.
Audit Metadata