spawn-ori-eval
Warn
Audited by Socket on Aug 2, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The purpose is coherent for an eval skill, but the trust model is weak: it installs a remotely scripted, non-registry CLI and then forwards credentials and repo-derived data into that external binary. Given the limited provenance evidence for `ori`, this creates high supply-chain and credential-forwarding risk even without proof of malicious exfiltration.
Confidence: 83%Severity: 84%
Audit Metadata