agent-ui
Warn
Audited by Socket on May 14, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The core purpose is coherent, and the API key/proxy flow generally matches an agent UI product. However, the skill expands trust through remote registry code installation and explicit transitive skill installation, with an unverifed `belt-sh/cli` path not clearly documented as the official install route. Risk is driven more by supply-chain and transitive trust than by confirmed malicious behavior.
Confidence: 86%Severity: 68%
Audit Metadata