9router-chat
Warn
Audited by Socket on May 12, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s purpose is coherent as a router client, and the docs/source relationship appears legitimate, but its core design routes prompts and optional credentials through an arbitrary intermediary endpoint instead of official model APIs. That makes the data flow materially higher risk than a direct OpenAI/Anthropic integration, especially because NINEROUTER_URL can point anywhere.
Confidence: 89%Severity: 74%
Audit Metadata