skills/akunzai/agent-skills/mem-auto/Gen Agent Trust Hub

mem-auto

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill includes a dedicated security specification in references/security.md that strictly prohibits the recording of authentication secrets, private keys, and personally identifiable information (PII).
  • [COMMAND_EXECUTION]: The memory autopilot utilizes a shell script (mem-sync-git.sh) for Git-based synchronization of project logs, which is a documented and standard function of the skill for cross-device memory consistency.
  • [SAFE]: Data management is handled through structured markers (e.g., [Handoff], [Candidate]) and dedicated sub-skills, providing clear separation between transient session state and long-term durable memory.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 11:36 PM
Security Audit — agent-trust-hub — mem-auto