1337
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external artifacts, providing an ingestion surface for indirect prompt injection.
- Ingestion points: Artifacts, logs, samples, and tool outputs as specified in SKILL.md and references/brain.md.
- Boundary markers: The skill identifies the risk and recommends loading a separate untrusted-input-hygiene reference skill.
- Capability inventory: The agent is instructed to perform analysis tasks and write state to a local brain.md file or task notes.
- Sanitization: Sanitization of external content is addressed by a separate recommended reference skill.
Audit Metadata