memory-personalization

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill establishes a mechanism for persisting and retrieving user-influenced data, which creates a potential surface for indirect prompt injection.
  • Ingestion points: User preferences and facts are processed via recall_memories and session context as described in SKILL.md.
  • Boundary markers: No specific delimiters or 'ignore' instructions for memory retrieval are specified in the instructions to separate stored data from system instructions.
  • Capability inventory: The agent uses the save_memory and recall_memories tools to manage its persistent state, allowing it to modify and read stored information.
  • Sanitization: The skill includes instructions to exclude health, financial, and identity details from stored memory, which acts as a safeguard against the persistence of sensitive data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 12:15 AM