venice-models

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill performs runtime fetches to Venice API endpoints (https://api.venice.ai/api/v1/models?type=text, https://api.venice.ai/api/v1/models/traits?type=text, https://api.venice.ai/api/v1/models/compatibility_mapping?type=text) and uses the returned model_spec/constraints to choose models and validate prompts, so the external responses directly influence agent prompts/behavior.