blog-refresher
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to use a page fetcher to audit the content of top-ranking competitor websites (Phase 2: Competitor research). This processing of untrusted external data creates a surface for indirect prompt injection, where instructions hidden on those pages could potentially influence the agent's output.
- Ingestion points: External competitor content fetched during Phase 2.
- Boundary markers: The instructions lack specific delimiters or "ignore instructions" warnings to wrap the external content during processing.
- Capability inventory: The skill is limited to content analysis and text generation; it does not utilize high-risk tools such as arbitrary shell execution or local file-system writes.
- Sanitization: No sanitization or validation of the fetched external text is performed before interpolation into the agent's context.
Audit Metadata