save-context
Warn
Audited by Socket on Mar 23, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the core snapshot-writing behavior is coherent, but the skill exceeds that purpose by autonomously committing and pushing saved conversation context to an arbitrary preconfigured git remote. Main risk is unintended disclosure of conversation/project details rather than malware or supply-chain abuse.
Confidence: 91%Severity: 72%
Audit Metadata