AgentChat-OneWeb
Warn
Audited by Socket on Oct 8, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill’s capabilities mostly fit its stated purpose, and the flagged scanner items are largely documentation artifacts or normal local diagnostics. However, it mandates immediate execution of a local script, forwards prompts and optional local images to multiple third-party AI services through logged-in browser sessions, and writes downloaded content to disk; these are significant but purpose-consistent risks rather than clear malware.
Confidence: 90%Severity: 58%
Audit Metadata