page-cro
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by utilizing the WebFetch tool to ingest external data from marketing pages. This could allow instructions on the analyzed pages to manipulate the agent's behavior.
- Ingestion points: Data from external websites is ingested via the WebFetch tool specified in the frontmatter.
- Boundary markers: The prompt provides no delimiters or specific instructions to help the agent distinguish between its core instructions and the fetched web content.
- Capability inventory: The skill relies on the Read and WebFetch tools for its operations.
- Sanitization: There is no logic within the skill to sanitize, escape, or filter retrieved content to prevent embedded instructions from being executed.
Audit Metadata