mvcc
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill content is purely informational, detailing database internals, file structures, and development testing workflows. No executable scripts, network exfiltration attempts, or malicious prompt injection patterns were found. The instruction to 'Ignore MVCC when debugging' is a benign scoping directive for focus, rather than an attempt to bypass agent safety guidelines.
Audit Metadata