bleu
Warn
Audited by Socket on Jun 13, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the core planning purpose is plausible, but the actual footprint is broader than a simple architecture skill. Mandatory web research, optional ingestion of untrusted external content, persistent file writes, hook/agent automation, and command-capable Claude Code integrations create medium risk through prompt-injection and workflow automation exposure, even without clear credential theft or malware behavior.
Confidence: 100%Severity: 60%
Audit Metadata