role-creator
Warn
Audited by Socket on Mar 22, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the file-generation behavior fits the stated purpose, but the skill's trust model is not proportionate. Its core workflow relies on an unverified external CLI (`agent-team`) and its generated roles are instructed to discover and install additional third-party skills from community ecosystems, creating a high transitive supply-chain risk. No confirmed malware or credential harvesting is shown, but the installation and execution footprint is broader and less trustworthy than a simple role-package generator should require.
Confidence: 85%Severity: 82%
Audit Metadata