memory-types
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a system for storing persistent memories that directly influence future agent behavior, specifically through the 'feedback' memory type.
- Ingestion points: Information is extracted from user messages and saved to the memory system based on the definitions in SKILL.md.
- Boundary markers: The instructions do not provide explicit delimiters or warnings to ignore embedded instructions when reading from the stored memory.
- Capability inventory: The system allows the agent to save 'feedback' which is explicitly used to guide future behavior and decision-making ('how_to_use' field).
- Sanitization: There is no requirement or guidance for sanitizing user-provided feedback before it is promoted to a persistent behavioral guideline.
Audit Metadata