deep-research
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process data from external web sources, which inherently introduces a surface for indirect prompt injection attacks.
- Ingestion points: External content is ingested via the
WebSearchandWebFetchtools during the research phase defined inSKILL.md. - Boundary markers: The instructions do not specify the use of clear delimiters (e.g., XML tags or triple quotes) or safety warnings to ignore instructions embedded within the fetched external content.
- Capability inventory: The skill instructions include reading from the web and writing structured reports to the
02-research/directory. - Sanitization: There is no evidence of explicit content sanitization or filtering logic to prevent the execution of instructions found in external data.
Audit Metadata