browser-visual-ux-qa
Warn
Audited by Socket on Oct 22, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The core browser QA and screenshot-analysis behavior aligns with the stated purpose, and the command-injection finding is a false positive from documentation code. The main risk is autonomy scope: the skill authorizes the agent to act without asking, modify environment state, commit/push, and deploy, which creates meaningful operational risk even though there is no clear malware or credential-harvesting behavior in the skill text itself. Third-party Context7 usage is coherent with the purpose but introduces moderate external data-flow and package-execution trust concerns if used via npx/API key.
Confidence: 91%Severity: 64%
Audit Metadata