session-review
Warn
Audited by Socket on Sep 1, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the review goal is plausible, but the skill's footprint extends into transitive skill trust and optional external model routing without clearly bounded provenance or data-flow controls. The main risk is delegating repo context/files to an ambiguously sourced reviewer that may also browse untrusted web content and then drive local changes.
Confidence: 83%Severity: 76%
Audit Metadata