on-page-seo-checker
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting HTML and content from external URLs (Ingestion points: SKILL.md, Step 1 and 9). The instructions include a boundary marker warning the agent to treat fetched content as untrusted data (Boundary markers: Present in SKILL.md). The skill's capabilities are limited to reading data and generating SEO reports (Capability inventory: WebFetch, onpage.py, firecrawl.py), and no specific sanitization or filtering of the external content is implemented (Sanitization: Absent). This finding is assessed as low severity.
- [COMMAND_EXECUTION]: The skill references local Python scripts (onpage.py, firecrawl.py, schema_lint.py) in the plugin root for metadata extraction. These are controlled internal helpers for the skill's SEO diagnostic functions and do not provide a path for arbitrary command execution.
Audit Metadata