planetscale-autonomous-execution-mode
Warn
Audited by Socket on Sep 4, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill’s purpose and capabilities are internally consistent and it uses official PlanetScale interfaces with no installer or credential-forwarding concerns. However, it deliberately enables autonomous production-affecting database actions after a one-time acknowledgment, which is inherently high-impact for an AI agent even with strong guardrails. The scanner’s 'hide actions' claim is not supported by the text; the main risk is autonomous execution, not covert behavior.
Confidence: 93%Severity: 68%
Audit Metadata