tao-launch-workflow
Shared launch intake for any TAO workflow or action. Use when the user wants to run TAO AutoML, train, evaluate, infer, export, generate TensorRT engines, or launch DEFT/workflow jobs on an execution platform.
npx skills add NVIDIA/skills --skill tao-launch-workflow --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Launch intake for any TAO workflow or action, coordinating preflight checks, platform selection, credentials, container image resolution, dataset inputs, compute shape, and monitoring preferences before launching a TAO job.
How it works
- Serves as a gate before launching any TAO workflow or model action.
- Forces a preflight process that ensures: execution platform selection, platform credentials and model credentials, default container image resolution (from metadata) with possible override via image=<override>, platform access, dataset input mapping to platform-specific spec keys, required compute shape fields, presence of local tools, and a launch review detailing image, platform, datasets, compute shape, and config changes.
- If any preflight item is missing, it prompts for input and stops before artifacts are generated.
- Guides the user through initial questions to select platform, monitoring preferences, and status cadence.
- Provides a structure for missing-input prompts with examples and both dataset input modes, and describes container image confirmation steps using resolve_tao_image.py when available.
- Includes a detailed platform preflight and dependency remediation flow, including handling of missing tools, path validation, and staged data considerations for large datasets.
When to use it
Use before launching any TAO workflow or model action, including AutoML, train, eval, infer, export, TRT, or DEFT/workflow launches, to ensure all preflight checks are satisfied and a safe launch plan is in place.
What it can touch
- Platform details and credentials prompts
- Container image resolution and confirmation
- Dataset inputs and platform-specific spec paths
- Preflight helper scripts: check_tao_launch_preflight.py, resolve_tao_image.py (via TAO skill-bank helpers)
Caveats
- Requires packaged TAO skill bank helper scripts
- Uses a multi-step preflight process; will stop and ask for missing inputs if any item is unsatisfied
- Monitoring behavior: final response ends chat-side monitoring; ongoing polling continues while running
- Overrides and tool usage must follow explicit commands and exact parameter formats as described
# TAO Workflow Launch Intake Use this skill before launching any TAO workflow or model action. ## Quick Start Run the platform helper, ask for platform and monitoring preferences, then run the selected platform detail helper before asking for credentials. ## Non-Negotiable Launch Gate This gate is model-agnostic. Apply it to every TAO model, data action, and application workflow before launching s
What does the tao-launch-workflow skill do?
Shared launch intake for any TAO workflow or action. Use when the user wants to run TAO AutoML, train, evaluate, infer, export, generate TensorRT engines, or launch DEFT/workflow jobs on an execution platform.
How do I install it?
Run `npx skills add NVIDIA/skills --skill tao-launch-workflow --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From NVIDIA/skills, a repository with 2,789 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.