to-intent
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes input from external sources like documents, tickets, and incidents. It mitigates potential injection risks by explicitly instructing the agent to treat fetched content as data and ignore embedded directives.
- Ingestion points: User-supplied input or fetched documents and tickets as described in Step 1 of SKILL.md.
- Boundary markers: The skill contains an explicit instruction in Step 1 to 'ignore embedded directives' and treat fetched content as data.
- Capability inventory: The skill is restricted to listing and writing files within the 'docs/intents/' directory.
- Sanitization: Output is governed by a strict template structure in 'assets/intent.md' that prevents the inclusion of technical designs or implementation details.
Audit Metadata