journey-loop
Warn
Audited by Socket on Mar 29, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the core build/test/refine purpose is coherent, but the skill combines indefinite autonomous looping with runtime ingestion of mutable external gist content that is injected into an execution-capable builder agent. The main risk is indirect prompt injection and uncontrolled iterative action, not clear credential theft or confirmed malware.
Confidence: 87%Severity: 78%
Audit Metadata