jsonl-digest
Warn
Audited by Socket on Apr 16, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
Mostly coherent local memory-digestion skill with no evident exfiltration or credential harvesting, but it relies on an unverifiable local executable script and performs autonomous file writes from transcript-derived content. Overall this is better classified as SUSPICIOUS/VULNERABLE due to supply-chain trust and transcript-to-write integrity risk, not as confirmed malware.
Confidence: 86%Severity: 72%
Audit Metadata