aws-ai
Warn
Audited by Socket on Mar 25, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill’s purpose and AWS-focused capabilities are broadly aligned, and the stated data flow appears to target AWS rather than an obvious interceptor. However, it asks the user to clone and run a lightly verifiable third-party helper container with direct access to `~/.aws` credentials and env-based role settings, which is a meaningful trust and credential-forwarding risk for an agent skill.
Confidence: 82%Severity: 62%
Audit Metadata