gpt-4o
Warn
Audited by Socket on Sep 22, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill's main inconsistency is branding itself as OpenAI Text-to-Speech while routing all requests and credentials through Pixazo instead of OpenAI's official API. The flow is coherent for a Pixazo gateway skill, but not for a direct OpenAI skill; local key persistence and the transitive skill-install instruction add moderate risk.
Confidence: 92%Severity: 72%
Audit Metadata