context-monitor
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a meta-instruction set providing the AI with patterns to recognize conversation drift (length, topic mixing, error loops). It suggests natural language strategies for refocusing, such as summarizing or suggesting task breakdown. No malicious command execution, data exfiltration, or obfuscation was found. The instructions are focused on maintaining conversation quality and do not attempt to bypass safety filters or extract sensitive information.
Audit Metadata