documentation-lookup
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (MEDIUM): The skill handles untrusted external documentation that could contain instructions to override the agent's behavior.
- Ingestion points: Content retrieved via the
query-docstool call inSKILL.md. - Boundary markers: Absent. No delimiters are defined to isolate untrusted documentation from the agent's instructions.
- Capability inventory: The skill influences agent reasoning and generates code examples for the user based on fetched data.
- Sanitization: Absent. No filtering or validation is performed on the documentation service's response.
Audit Metadata