guardian
Warn
Audited by Socket on May 6, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally consistent with its stated purpose as a team progress monitor, and there is no clear credential theft, external exfiltration, or supply-chain abuse. However, its footprint includes persistent cron-based autonomy, direct tmux input injection, automatic approval of prompts, and self-modification of its protocol file; these go beyond passive monitoring and create meaningful operational risk even though they still broadly fit the skill's goal.
Confidence: 89%Severity: 72%
Audit Metadata