natividad-zero-click
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data provided by the user, such as customer questions, support notes, interviews, and community discussions. 1. Ingestion points: SKILL.md (Step 3) instructs the agent to review external evidence to identify audience triggers. 2. Boundary markers: The skill does not define explicit delimiters for untrusted input in the ingestion step, although it includes a research honesty check in checklists/review-checklist.md to separate facts from assumptions. 3. Capability inventory: The skill's functionality is limited to content analysis and text generation; no dangerous capabilities such as shell execution, system-level file access, or network operations were found. 4. Sanitization: The instructions do not prescribe sanitization or escaping of external content before processing.
Audit Metadata