context-surfing
Warn
Audited by Socket on Apr 24, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The core behavior is coherent for a context-management skill and data flow stays local, but the install model creates transitive trust in a personal third-party skill repo, and the hook/CLI integrations add medium supply-chain and prompt-injection exposure. No clear credential harvesting, exfiltration, or malicious intent is present.
Confidence: 83%Severity: 58%
Audit Metadata