forge-plan
Warn
Audited by Socket on Apr 16, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
Mostly aligned orchestration skill, not overtly malicious. Main risk is that it feeds untrusted GitHub/local content into write- and execution-capable subskills with automatic phase progression, creating meaningful prompt-injection and autonomous code-change risk; tool provenance for several delegated skills is also unspecified.
Confidence: 83%Severity: 61%
Audit Metadata