memory-management
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute various Python scripts (e.g., query_cache.py, query.py, vault_ops.py) and command-line utilities like ripgrep (rg) and grep to perform searches and manage memory tiers.
- [DATA_EXFILTRATION]: The architecture includes a 'Soul' tier that facilitates the synchronization of session traces to external storage services like the HuggingFace Hub for long-term persistence, which is a core functional feature of the skill.
Audit Metadata