where-were-we
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes arbitrary conversation history to generate a recap for the user.
- Ingestion points: Conversational turns ingested by the agent as described in SKILL.md.
- Boundary markers: None identified; instructions do not include specific delimiters for the recap content.
- Capability inventory: No tool use, network access, file system modification, or code execution capabilities were identified in the skill.
- Sanitization: None identified; the skill recaps text as-is from history.
Audit Metadata