conversation-memory
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for 'Memory-Aware Prompting' where retrieved memories are included in agent prompts. This creates an attack surface where malicious instructions embedded in a user's previous conversation could be stored and later executed when retrieved and placed back into the context.
- Ingestion points: Processes retrieved conversation memories as described in SKILL.md.
- Boundary markers: No specific delimiters or 'ignore' instructions for retrieved content are specified.
- Capability inventory: The skill itself contains no scripts or tool definitions.
- Sanitization: No mention of sanitizing or validating retrieved memories before use.
- [NO_CODE]: The skill consists exclusively of markdown documentation. It contains no executable scripts, tool configurations, or command-line instructions.
Audit Metadata