lovable
Warn
Audited by Socket on Oct 5, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's purpose, permissions model, and data flow are mostly coherent with an official Lovable integration, and it avoids asking for secrets in chat. The main issue is install/execution trust: it relies on an external `lovable` CLI whose provenance and release verification are not established in the skill, so the binary remains a black box capable of receiving project data and invoking remote actions.
Confidence: 85%Severity: 72%
Audit Metadata