memory-defrag
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only instructional text for the agent to perform data organization tasks. No malicious patterns, obfuscation, or unauthorized access attempts were detected.
- [DATA_EXPOSURE]: While the skill interacts with internal memory files (e.g., MEMORY.md and the memory/ directory), it does not include instructions for exfiltrating this data or accessing sensitive system files outside the agent's defined memory scope.
- [COMMAND_EXECUTION]: The skill does not contain shell commands, scripts, or instructions for the agent to execute arbitrary code. Mentions of 'cron' are framed as recommendations for the user rather than automated persistence mechanisms.
Audit Metadata