Lens
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided task descriptions to drive its analysis and web search functionality, creating an attack surface for indirect instructions.
- Ingestion points: User-provided task descriptions are accepted as input in SKILL.md.
- Boundary markers: The prompt instructions do not specify the use of delimiters or boundary markers to isolate user-supplied task descriptions from the system instructions.
- Capability inventory: The skill employs WebSearch and WebFetch tools to perform targeted research based on user-supplied parameters (e.g., tool names, people, or industries).
- Sanitization: No input validation or sanitization is performed on the user's task description or the content retrieved from external search results.
Audit Metadata