review
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing untrusted external data, which inherently creates an indirect prompt injection surface.
- Ingestion points: The skill accepts data through file paths, cloud document links, system IDs, and pasted text from the user.
- Boundary markers: The instructions do not define specific delimiters or 'ignore embedded instructions' markers for the content being processed.
- Capability inventory: The skill's capabilities are focused on document routing and generating review summaries; no dangerous shell execution or network exfiltration capabilities were identified.
- Sanitization: The skill does not specify any sanitization or validation logic for the ingested agreement text.
- [SAFE]: The skill's behavior is consistent with its stated purpose of assisting in legal document review. It does not utilize obfuscation, hardcoded credentials, or unauthorized persistence mechanisms. The inclusion of a 'confirm_routing' logic provides an optional human-in-the-loop verification step before sub-skills are executed.
Audit Metadata