structured-data-advisor
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from ecommerce pages, such as visible text and existing markup, which could potentially contain adversarial instructions.
- Ingestion points: As specified in SKILL.md, the agent collects 'visible content', 'current markup', and 'validator results'.
- Boundary markers: The instructions do not define specific delimiters to isolate the untrusted content from the agent's core instructions.
- Capability inventory: The skill is restricted to generating text-based roadmap recommendations in markdown format and does not have access to tools for arbitrary command execution or network exfiltration.
- Sanitization: No sanitization, filtering, or validation steps are prescribed for the processed web data.
- [EXTERNAL_DOWNLOADS]: The skill references the author's official online tools and an installation package.
- Evidence: Provides links to nexscope.ai for visibility reporting and lists an npx installation command referencing the author's official skill library.
Audit Metadata