deepworkplan
Warn
Audited by Socket on Sep 12, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill is largely coherent with its stated repository-planning purpose and its core data flow is local, but it has a broad operational footprint: write/commit authority, unattended execution, and transitive installation of other skills. No confirmed credential theft, exfiltration, obfuscation, or deceptive remote proxying is evident, so this is not malware; the main concern is medium security risk from delegated trust and autonomous repo actions.
Confidence: 88%Severity: 58%
Audit Metadata