dynamo-recipe-runner

Validate and patch existing Dynamo Kubernetes recipes, then deploy and smoke-test endpoints.

Updated Apr 16, 2026

Search Tags:#validation #kubernetes #recipes #deploy #bring-up #dynamo

Ready to run

npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill dynamo-recipe-runner

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dynamo-recipe-runner
Source: https://github.com/sayalinvidia/sayali-skills-test/tree/main/skills/dynamo-recipe-runner
Command: npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill dynamo-recipe-runner

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Select, validate, patch, and deploy existing NVIDIA Dynamo Kubernetes recipes to bring up model/backend/GPU/deployment-mode environments with minimal manual steps and consistent results.

Core Features & Use Cases

Discover available Dynamo recipes and validate them with the provided tooling.
Patch only cluster-specific values (storage class, image tag, Hugging Face secret, GPU count) and deploy model-cache then deploy.yaml.
Smoke-test the frontend endpoint and verify a healthy DynamoGraphDeployment in the target namespace.

Quick Start

Run the recipe_tool to list Dynamo recipes, select a matching one, validate it, patch only cluster-specific values, apply model-cache and deploy.yaml, and smoke-test the frontend endpoint.

Frequently Asked Questions about dynamo-recipe-runner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deploy NVIDIA Dynamo Kubernetes recipes to a cluster?▼

To deploy Dynamo Kubernetes recipes, you can validate existing recipes, patch cluster-specific values like storage class and image tag, apply model-cache and deploy.yaml manifests, and smoke-test the endpoint to ensure a healthy deployment.

What is the process for validating Dynamo recipes before deployment?▼

Validating Dynamo recipes involves identifying available recipes and using provided tooling to verify them, ensuring they match your model, backend, GPU, and deployment-mode environment before applying any patches.

Can I create new Kubernetes manifests when bringing up Dynamo environments?▼

No, you cannot create new manifests. Safe operation requires using only existing Dynamo recipes and preparing minimal, cluster-specific patches while reporting any blockers or outcomes encountered.

How do I patch cluster-specific values for a Dynamo deployment?▼

Patching cluster-specific values involves updating only storage class, image tag, Hugging Face secret, and GPU count in the existing Dynamo recipes to match your target environment before applying manifests.

Does the Dynamo recipe runner support smoke-testing frontend endpoints?▼

Yes, after deploying model-cache and deploy.yaml, it smoke-tests the frontend endpoint and verifies a healthy DynamoGraphDeployment in the target namespace to confirm readiness.

What should I do if readiness signals fail during Dynamo recipe deployment?▼

If readiness signals fail during deployment, the recipe runner reports blockers and outcomes, ensuring you are alerted to issues preventing a healthy DynamoGraphDeployment in your namespace.