Agent skill · nvidia
paidf-orchestration-write-dag
Use when a user describes a custom PAIDF Orchestration pipeline — a specific ordered combination of stages such as augmentation only, auto-labeling only, detection+captioning only, or image attribute augmentation without full auto-labeling — that no existing DAG in airflow/dags/workflows/ covers, and asks for a new Kubernetes DAG. Also use to check that a generated or existing DAG's model/container/prompt choices match an external spec document (e.g. a PAIDF `launchable.md`).
What it needs
About 10k tokens when loaded.
What this skill does
Write DAG What This Skill Does Composes a new Kubernetes-only Airflow DAG from the existing, shared task groups in airflow/dags/shared/taskgroups/ — the same building blocks imageattributeaugmentationdag.py and eventvideogenerationdag.py are built from. It does not invent new task-group logic. Output is: 1. airflow/dags/workflows/<dagname>dag/configs/<dagname>k8smanifest.yaml — compute + component manifest 2. airflow/dags/workflows/<dagname>dag/<dagname>dag.py — the DAG builder 3. airflow/dags/workflows/<dagname>dag/models/payload.py — the payload pydantic model 4. airflow/dags/workflows/<dagname>dag/callables/ — required workflow-local glue, not optional. Every TaskGroup used here takes one or more Callable constructor args with no default (e.g. InputPreparationTaskGroup(prepareinputcallable=...), CosmosTaskGroup(configgenerationcallable=..., outputvalidationcallable=...), ValidatedOutputTaskGroup(validationcallable=...)). These callables are genuinely new, per-DAG code — they are what turns generic container-arg building into this pipeline's actual input listing / config generation / output validation. Do not skip writing them, and do not treat "don't invent task logic" as license to invent them freely either: for each task group you use, find the closest existing DAG that uses the same task group, read its equivalent callable, and adapt it — trim the parts that don't apply (e.g. dropping an outputgroupid/downstream-XCom concern when there is no next stage) rather than writing from a blank page. Kubernetes only. This repository's main/current branches check in K8s manifests only (see image-attribute-augmentation-workflow and event-video-generation-workflow skills) — no NVCF, no OSMO. Do not offer those backends unless the user explicitly says their environment supports them, and if so, treat it as new scope requiring its own investigation rather than a flag on this skill. Not scoped to one workflow. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills nvidia/paidf-orchestration-write-dag