BackstopJS Visual Regression Testing Automation
Warn
Audited by Socket on Apr 6, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The stated purpose aligns with visual regression testing and the referenced BackstopJS project is real and well-scoped, with no obvious credential harvesting or exfiltration. The main concern is transitive installation: the skill asks the agent to install a third-party skill from `agentskillexchange/skills` rather than install/use BackstopJS directly, which adds supply-chain and inherited-permission risk disproportionate to a simple framework guide. Overall this looks coherent but medium-risk due to the extra trust layer.
Confidence: 84%Severity: 56%
Audit Metadata