fal-recipes
Warn
Audited by Socket on Sep 14, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill’s media-generation capabilities align with its stated purpose, and the network/data flows mostly match expected fal.ai workflows. The main risk is trust expansion: it depends on an external CLI installed via pipe-to-shell, forwards API credentials through that CLI, and uses `genmedia init` for transitive skill loading. That is not fundamentally incompatible with the purpose, but it is broader and riskier than a purely local documentation skill.
Confidence: 87%Severity: 54%
Audit Metadata