omniroute-integration

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly prompts for the OmniRoute API key and describes updating auth.json and making test API calls, which requires the agent to embed and therefore output the secret verbatim into configuration or requests.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill's required workflow (skill.yml and the included install/test scripts) configures a remote OmniRoute endpoint (default http://138.2.29.190:20128) and explicitly runs a test model call (opencode run -m omniroute/kr/claude-sonnet-4.5 ...) which consumes and displays arbitrary responses from third‑party AI models, so untrusted external content is read as part of the runtime verification step.