tao-run-deft-aoi-cosmos3

tao-run-deft-aoi-cosmos3

Installation

Install this application as part of the full TAO skill-bank root, not as only the companion skill folders: a usable install places skills/, scripts/, templates/, and versions.yaml under the same TAO_SKILL_BANK_PATH. Any install that ships only the skill folders, for example an agent plugin or skills-only install, must also provide the bank-level scripts, templates, and versions.yaml and point TAO_SKILL_BANK_PATH at their common root before running scripts/resolve_tao_model.py. Run bundled validation with the skill Python so dependencies match runtime: PYTHON=$(bash scripts/deft_python.sh); "$PYTHON" -m unittest tests.test_cosmos3_bare. Resolve network mode first. Missing air-gap imports are a hard stop; network-enabled setup lives only in references/network-bootstrap.md.

Execution Contract

Treat a run as a disk-backed state machine.

1. Preserve every explicit user value and show the source of each effective value (user, spec, or default) in the Pre-Flight Summary.
2. Ask which installed platform to use. Do not select Docker, SLURM, Kubernetes, Brev, virtualenv, or an external platform by default.
3. Resolve network mode, then read exactly one of references/air-gap.md or references/network-bootstrap.md. Run the selected platform skill's Preflight and stop on a missing system/native-CLI prerequisite.
4. Before any mutation or launch, invoke tao-launch-workflow and show its launch review plus this skill's Pre-Flight Summary. Wait for one explicit approval.
5. After approval, set PYTHON=$(bash scripts/deft_python.sh) and initialize ${RESULTS_DIR}/deft_state.json once with "$PYTHON" scripts/init_deft_state.py. Require --base-model-path to name the prepared Qwen3-VL PTM, freeze the image with --framework-image-digest, and pass its container URI with --framework-container. Pass the exact GPU model reported by the selected platform's Preflight through --gpu-model (include accelerator memory when available), plus the resolved network mode/source and selected absolute Python. Never reinitialize a resumed run or edit deft_state.json by hand.
6. Before every stage, after context compaction, and before a completion claim, run "$PYTHON" scripts/deft_context.py --state ... --stage .... Use its durable next_stage and the state file's status, current_iteration, iterations.*.status, stage_completed, and latest events entry to resume. Do not infer progress from assistant prose or from an artifact that is not recorded in state.
7. Run every command that can install, fetch, log in, or launch a local container through "$PYTHON" scripts/deft_exec.py --state ... -- <command>. In an air-gap it rejects egress/package operations and enforces no-pull. Remote platforms must apply the equivalent immutable no-pull/offline policy.
8. Submit each GPU stage through the chosen platform's four verbs: submit / status / logs / cancel. The submit verb must open the job-record before native launch; the returned id is the only launch handle. Poll the backend, not the job-record, and map state to PENDING RUNNING COMPLETE ERROR CANCELED UNKNOWN.
9. Commit every completed or failed DEFT stage with "$PYTHON" scripts/commit_stage.py. It verifies the stage inputs and atomically updates both the resume snapshot and ordered events array in state. commit_stage.py --stage train requires --framework-config with the saved Hydra config.yaml. Every executed-stage commit requires a positive, measured --duration-sec: use backend elapsed wall time for submitted jobs and a host wall-clock timer for inline stages. A documented --skip may record 0; negative durations are always rejected.
10. Claim completion only after "$PYTHON" scripts/finalize_run.py verifies final Benchmark evidence, successfully commits loop_stop, and a fresh read of deft_state.json shows status == "complete", version == 6, non-empty final_artifacts, iterations.baseline.status == "complete", and the final iteration's status == "complete".