llm-router
Warn
Audited by Socket on Apr 13, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's purpose is plausible, but its real footprint depends on an external CLI that may be auto-installed and then trusted with multiple API keys and prompts, while also exposing agentic MCP/tool capabilities beyond simple routing. The biggest issue is missing install provenance for SwiftOpenAI-CLI in the supplied material; without verified same-org official distribution details, the credential-forwarding and supply-chain risk are high.
Confidence: 77%Severity: 83%
Audit Metadata