archeology
Warn
Audited by Socket on Aug 31, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS due to scope breadth, recursive autonomous source expansion, and prompt-injection exposure from processing many untrusted content sources with write capability. It is not overtly malicious: there are no installs, no credential-harvesting steps, no proxy gateways, and no outbound exfiltration instructions. Main risk is disproportionate research scope for a single skill and indirect prompt injection during evidence gathering.
Confidence: 86%Severity: 58%
Audit Metadata