gpt-image-playground
Warn
Audited by Socket on Sep 14, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill largely matches its stated purpose and uses standard install paths, but it normalizes sending OpenAI API keys and uploaded images to arbitrary proxy URLs and even via URL query parameters. That data-flow flexibility is disproportionate enough to raise concern, though not to malicious levels.
Confidence: 91%Severity: 58%
Audit Metadata