ingest
Warn
Audited by Socket on Aug 29, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill's core behavior is mostly aligned with its stated ingestion purpose, but it persistently captures user and fetched content, uploads raw materials to external storage, and relies on a same-org but weakly verified external CLI/install chain. This looks more like a high-trust knowledge-ingestion skill than malware, but it carries meaningful privacy, supply-chain, and autonomous-write risk.
Confidence: 84%Severity: 52%
Audit Metadata