ai-memory-developer
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation providing design patterns and workflows for AI memory management. No executable scripts, binaries, or configuration files are included in the skill package.
- [DATA_EXFILTRATION]: The implementation checklists explicitly advise against storing sensitive data such as secrets, payment information, or full medical records unless specifically required and approved. It also mandates multi-tenant isolation (filtering by user_id and tenant_id) for all memory operations.
- [PROMPT_INJECTION]: The retrieval workflow documentation recommends the use of structured XML boundary markers and explicit model instructions to mitigate potential indirect prompt injection risks associated with memory retrieval.
Audit Metadata