zcode-delegate
Warn
Audited by Socket on Sep 2, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s purpose is coherent, and credential/network flows mostly target official Z.AI infrastructure, but the trust boundary is weak because the actual relay runtime is missing, the publisher is a community account, and the executed CLI path can be redirected to any local binary. This is not confirmed malware, but it is a high-risk delegation skill with meaningful supply-chain and credential-forwarding exposure.
Confidence: 86%Severity: 74%
Audit Metadata