sanity-check
Warn
Audited by Socket on Oct 15, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the stated purpose is coherent, but the skill implements it with load-time pre-execution. Although there is no credential theft or remote payload, running `make lint` and `go test` before approval executes repository-controlled code automatically, creating a high security risk without clear malicious intent.
Confidence: 91%Severity: 78%
Audit Metadata