agents-build
Warn
Audited by Socket on Sep 18, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
Mostly coherent AWS AgentCore build guidance, with many scanner hits attributable to documentation. The major security issue is scope: the payments sub-skill enables autonomous spending and introduces plaintext secret storage plus third-party frontend installation, making the overall skill medium/high risk even without clear malicious intent.
Confidence: 87%Severity: 71%
Audit Metadata