agent-memory-systems

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill content is purely educational and architectural, describing patterns for memory retrieval, chunking, and vector store selection. No malicious instructions or patterns were identified.
  • [NO_CODE]: The skill consists entirely of a Markdown file (SKILL.md) for documentation purposes. It does not contain any executable scripts (Python, Node.js, Shell), tool configurations, or environment modifications.
  • [METADATA_POISONING]: Metadata fields such as name, description, and source are descriptive and consistent with the architectural content of the skill. No deceptive patterns were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 02:31 PM
Security Audit — agent-trust-hub — agent-memory-systems