deploy-collection

End-to-end deployment of Content Agents is automated by orchestrating service startup, environment wiring, and plan generation through a single scripted workflow via the deploy.py script. It coordinates Material, Physics, and Texture Agents with optional shared dependencies.

Yes, you can deploy Content Agents together or in combos across CPU-only environments. The orchestration workflow validates environment configuration and wires shared endpoint dependencies for VLM, LLM, image generation, and embeddings.

The best way to orchestrate Content Agents startup without Helm is using the provided deploy.py script. It handles plan generation, environment wiring, and service startup for CPU-based deployments without requiring Helm charts.

Yes, the deployment workflow supports shared OVRTX rendering and wires endpoint dependencies for VLM, LLM, image generation, and embeddings. It also supports provider-neutral deployment with optional Brev-hosted dependencies.

Environment validation during Content Agents deployment ensures configuration is correct before service startup. The scripted workflow validates environment variables and wires endpoint dependencies automatically as part of the deployment plan generation.

The deployment workflow is designed specifically for CPU-based deployments without Helm. It orchestrates end-to-end deployment across multiple services but does not address GPU-based deployments or Helm-managed Kubernetes environments.