Agent skill · software engineering · nvidia
tao-run-deft-aoi
Run the full DEFT AOI improvement loop for NVIDIA TAO VisualChangeNet / ChangeNet PCB inspection models: baseline evaluate, RCA, Cosmos AnomalyGen / AMP synthetic defects, k-NN mining, retraining, and deployment gating against a customer-defined primary metric and optional constraints. Use only when the request identifies an AOI / automated-optical-inspection, PCB-defect, VisualChangeNet, or ChangeNet workflow. Supports air-gapped/offline runs with pre-staged assets. Never infer AOI from generic iterative-improvement language. Do not use for CLIP / SigLIP image retrieval, attribute-labelled image data, standalone TAO training, one-off inference, generic anomaly generation, or RCA-only analysis.
Why this skill is useful
Adds a comprehensive automation script for running the DEFT AOI improvement loop with specific NVIDIA TAO model configurations and dependencies.
What it needs
Requires docker, nvidia-container-toolkit installed locally. Requires huggingface account access. About 11k tokens when loaded. Last updated 2026-08-07. 2,816 stars on the source repository.
What this skill does
Skill: tao-run-deft-aoi Standalone install? If this session was not initialized by the TAO skill bank plugin, run the tao-setup skill first (host preflight, credentials, cross-skill discovery). Execution Contract Treat this as a disk-backed state machine, not as a prose recipe. 1. Preserve every explicit user value. epoch 1 means numepochs=1 and iteration 1 means maxiterations=1; a heuristic or spec default applies only when the user did not supply that parameter. Show the source of every run parameter (user, spec, or default) in the Pre-Flight Summary. Preserve the customer's metric name, operator, target, unit, evaluator, and constraints. The approved metriccontract is the source of truth for evaluation, checkpoint selection, completion, and reporting. 2. After the user approves the Summary, set PYTHON=$(bash scripts/deftpython.sh), then initialize deftstate.json once with "$PYTHON" scripts/initdeftstate.py, passing Preflight's exact GPU model/memory, resolved --network-mode, activation source, and selected absolute Python. The resulting executionpolicy is immutable run state. Never hand-author or reinitialize it on resume. 3. On startup, after context compaction, before every stage, and before any completion claim, run "$PYTHON" scripts/deftcontext.py --state ... --stage .... Use its durable nextstage plus the state file's status, currentiteration, iterations..status, stagecompleted, and latest events entry to resume. Do not infer progress from assistant prose or from an artifact that is not recorded in state. 4. Invoke the mapped underlying skill after reading the DEFT overlay. Do not replace a missing/unread stage reference or a failed skill call with guessed shell commands, inline Python, a different output tree, or data fabricated from the KPI set. 5. After initialization, run install/fetch/login/container commands through "$PYTHON" scripts/deftexec.py --state ... -- <command>. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills nvidia/tao-run-deft-aoi