results
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from external sources, which presents a standard surface for indirect prompt injection.
- Ingestion points: Data is fetched via
ask_revnuandresults_getfrom external channels including email replies, LinkedIn messages, and CRM records as described inSKILL.md. - Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" wrappers for the data fetched from external tools.
- Capability inventory: The skill allows the agent to call tools for reading pipeline, deals, and channel activity, but instructions explicitly state these are read-only and should not perform destructive actions like spending or publishing.
- Sanitization: No specific sanitization or filtering logic is described for the incoming tool data.
- [SAFE]: The skill functionality aligns with its stated purpose of growth reporting. All referenced domains (revnu.com) and tool prefixes are associated with the vendor, and no suspicious network or file system operations were found.
Audit Metadata