agent-context-engineering
Warn
Audited by Socket on Sep 25, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The stated purpose is coherent and the text includes sensible safety cautions, but the skill materially depends on unverified local executables whose provenance and release path are not disclosed. No clear credential theft or exfiltration is present, yet the undocumented CLI trust chain makes the skill high risk to execute as written.
Confidence: 84%Severity: 78%
Audit Metadata