autoresearch
Warn
Audited by Socket on May 23, 2026
2 alerts found:
AnomalySecurityAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The excerpt is internally coherent as an autonomous research/iteration skill, but its scope is broad and includes deployment orchestration and security/red-team review, which raises medium risk for autonomous real-world actions. No install chain, credential harvesting, or exfiltration path is present in the provided text, so this is not confirmed malware.
Confidence: 100%Severity: 60%
Securityautoresearch.md
MEDIUMSecurityMEDIUM
autoresearch.md
Audit Metadata