# tao-run-deft-aoi

Runs an improvement cycle for NVIDIA TAO VisualChangeNet and ChangeNet PCB inspection models, from baseline evaluation to deployment checks against a target metric.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1600 installs · ★ 3735
- Author: NVIDIA Corporation
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/NVIDIA/skills/tree/HEAD/skills/tao-run-deft-aoi/SKILL.md
- Updated: 2026-10-23
- HTML: https://affpapa.org/en/nevibecode/skills/tao-run-deft-aoi

## What it does

This skill coordinates a full AOI improvement cycle for PCB inspection models. It evaluates the baseline, investigates failure causes, creates synthetic defects with Cosmos AnomalyGen or AMP, mines data with k-NN, retrains the model, then checks deployment conditions against a customer-defined metric and constraints. Its scope is VisualChangeNet and ChangeNet workflows, not a standalone training run or one-off inference.

Progress is recorded in `deft_state.json`, which the workflow uses to resume from committed stages. Docker and `nvidia-container-toolkit` are required, and individual stages may need additional resources. Air-gapped runs rely on assets staged in advance; network access and installs are blocked in that mode. If the TAO skill bank plugin did not initialize the session, run `tao-setup` first.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add NVIDIA/skills --skill tao-run-deft-aoi -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/tao-run-deft-aoi/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/tao-run-deft-aoi/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add NVIDIA/skills --skill tao-run-deft-aoi -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/tao-run-deft-aoi/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add NVIDIA/skills --skill tao-run-deft-aoi -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/tao-run-deft-aoi/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install NVIDIA/skills tao-run-deft-aoi
```

2. **Manually: put the files in place**
- File: `.github/skills/tao-run-deft-aoi/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add NVIDIA/skills --skill tao-run-deft-aoi -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/NVIDIA/skills --path skills/tao-run-deft-aoi
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/tao-run-deft-aoi/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add NVIDIA/skills --skill tao-run-deft-aoi -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/NVIDIA/skills/tree/HEAD/skills/tao-run-deft-aoi/SKILL.md

## When to use it
- evaluate a baseline VisualChangeNet or ChangeNet model
- trace failures in PCB inspection results
- generate synthetic defects for model training
- retrain toward a target evaluation metric
- check a model against deployment conditions

## Example prompts
- Run the DEFT improvement cycle for my ChangeNet AOI model and work toward the target metric.
- Evaluate my VisualChangeNet baseline, investigate inspection failures, and continue the AOI improvement workflow.
- Set up the PCB inspection workflow for an air-gapped run with assets staged in advance.

## Similar
- [skill-creator](https://affpapa.org/en/nevibecode/skills/skill-creator.md) — skill · 414180 installs
- [writing-for-agents](https://affpapa.org/en/nevibecode/skills/writing-for-agents.md) — skill · 466719 installs
- [media-use](https://affpapa.org/en/nevibecode/skills/media-use.md) — skill · 1100000 installs
- [mcp-builder](https://affpapa.org/en/nevibecode/skills/mcp-builder.md) — skill · 130949 installs
- [subagent-driven-development](https://affpapa.org/en/nevibecode/skills/subagent-driven-development.md) — skill · 237746 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 349800 installs

From the same set: [nvidia/skills](https://affpapa.org/en/nevibecode/by/nvidia/skills.md) — 194 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md