Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NVIDIA-TAO/tao-skill-bank --skill tao-run-deft-pasgit clone --depth 1 https://github.com/NVIDIA-TAO/tao-skill-bankWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-run-deft-pas)<a href="https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-run-deft-pas"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-run-deft-pas/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-run-deft-pas"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-run-deft-pas.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 146 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00175 | $0.04363 |
| Opus 5.5 | $0.00070 | $0.01745 |
| Sonnet 5.5 | $0.00035 | $0.00873 |
| Haiku 4.5 | $0.00017 | $0.00436 |
Grade A, and why
tao-run-deft-pas scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 15d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 362 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PAS DEFT Workflow
Standalone install? If this session was not initialized by the TAO skill bank plugin, run the
tao-setupskill first for host preflight, credential checks, and cross-skill discovery.
Run the canonical PAS flow as one resumable, disk-backed workflow. All PAS workflow logic, templates, and host adapters ship with this skill; customers do not need a separate source checkout. The bundled scripts make stage calls deterministic without adding another orchestration layer.
This skill supports every packaged TAO execution platform: Docker, SLURM,
Kubernetes, Brev, and virtualenv. The workflow produces a platform-neutral,
schema-validated action bundle; the selected platform skill owns native
submit/status/logs/cancel and the job-record.
Virtualenv execution uses separate immutable pyt and ds runtime profiles;
the workspace control .venv is not an execution runtime.
Entry Contract
The archive-location boundary is active as soon as the user supplies a data
root, archive root, or archive file path. The first and only operation allowed
to enumerate or inspect that location's contents is the packaged
scripts/discover_pas_inputs.py invocation. Before it runs, do not use Bash,
Read, Glob, list-directory tools, or agent-authored Python to inspect the
location. After it runs, do not perform a second search. This is a capability
boundary for the entire intake, not an ordering suggestion inside preflight.
tao-deft-pas and tao-run-deft-pas select this same workflow. If the user
did not choose a platform, ask once in plain text: Execution platform: Docker, SLURM, Kubernetes, Brev, and virtualenv? Never default or use an option-limited
control. On resume, the immutable platform in
deft_state.json is already the selection and must not be changed.
Use two intake phases:
- Perform only bounded, lightweight path discovery. Two explicit archive file
paths win and suppress all archive searching. An explicit archive directory
is treated as one approved search root. Otherwise, resolve the conventional
~/workspaceand use only these deduplicated search roots:~/pas, the workspace root, and itspas/,input/, andinputs/children. Check the workspace root itself only; beneath each other search root, inspect the root and directories at most two levels below it. This AOI-style bounded-subtree lookup supports nested export/drop directories without turning into a home or repository scan. Never follow symlinks or add the current checkout, source repositories, tutorial/notebook trees, or workspace dataset-output trees as implicit search roots.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- .env.example 1.2 KB
- eval.config 3.3 KB
- evals/evals.json 31 KB
- patches/sitecustomize.py 2.5 KB runs code
- references/action-request.schema.json 4.8 KB
- references/clip-train-eval.md 9.2 KB
- references/data-layout.md 6.0 KB
- references/gap-analysis.md 2.3 KB
- references/job-binding.schema.json 1.3 KB
- references/metric-contract.md 3.8 KB
- references/mining.md 6.7 KB
- references/pipeline-and-state.md 7.9 KB
- references/platform-action-status.schema.json 2.6 KB
- references/platform-execution.md 14 KB
- references/preflight.md 41 KB
- references/scripts-and-agents.md 14 KB
- references/staging-receipt.schema.json 1.1 KB
- references/virtualenv-runtime-manifest.json 4.6 KB
- references/virtualenv-runtime-manifest.schema.json 3.5 KB
- references/visualization.md 5.8 KB
- scripts/audit_deft_run.py 142 KB runs code
- scripts/check_pas_control_prereqs.py 6.0 KB runs code
- scripts/check_pas_credentials.py 3.2 KB runs code
- scripts/check_pas_cuda_runtime.py 7.4 KB runs code
- scripts/checkpoint_contract.py 9.1 KB runs code
- scripts/classify_pas_image_acquisition.py 8.0 KB runs code
- scripts/command_contract.py 9.5 KB runs code
- scripts/commit_stage.py 57 KB runs code
- scripts/deft_action_contract.py 21 KB runs code
- scripts/deft_python.sh 3.8 KB runs code
- scripts/discover_pas_inputs.py 10 KB runs code
- scripts/finetuning_contract.py 4.0 KB runs code
- scripts/init_deft_state.py 32 KB runs code
- scripts/log_stage.py 4.8 KB runs code
- scripts/manage_pas_virtualenv.py 4.1 KB runs code
- scripts/metric_contract.py 14 KB runs code
- scripts/parse_pas_metrics.py 7.6 KB runs code
- scripts/pas_deft/__init__.py 480 B runs code
- scripts/pas_deft/analyze_gaps.py 45 KB runs code
- scripts/pas_deft/config_fields.py 2.2 KB runs code
- scripts/pas_deft/config.py 35 KB runs code
- scripts/pas_deft/data_mining.py 69 KB runs code
- scripts/pas_deft/dataset_layout.py 4.8 KB runs code
- scripts/pas_deft/history_aware_mining.py 25 KB runs code
- scripts/pas_deft/pairs_io.py 1.8 KB runs code
- scripts/pas_deft/pas_artifacts.py 551 B runs code
- scripts/pas_deft/utils.py 27 KB runs code
- scripts/pas_deft/visualization.py 26 KB runs code
- scripts/path_contract.py 1.3 KB runs code
- scripts/prepare_deft_config.py 25 KB runs code
- scripts/record_metric_result.py 8.2 KB runs code
- scripts/recover_commit.py 4.2 KB runs code
- scripts/render_deft_report.py 27 KB runs code
- scripts/run_deft_action.py 49 KB runs code
- scripts/run_deft_cli.py 16 KB runs code
- scripts/run_deft_container.py 28 KB runs code
- scripts/run_deft_docker_action.py 13 KB runs code
- scripts/run_pas_runtime_probe.py 5.5 KB runs code
- scripts/run_pas_stage.py 39 KB runs code
- scripts/virtualenv_runtime.py 13 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 15d ago Changed · +15 lines 02c9f552f5a9
- 18d ago Changed · +15 lines 826d4a98803e
- 19d ago Changed · +1 lines 7d4cb3c9cf05
- 29d ago Changed · +14 lines ad4f75d7b660
- 1mo ago First seen · 317 lines · 175 tokens per session scan A 8b00d15c8c09
tao-run-deft-pas is a skill published in the GitHub repository NVIDIA-TAO/tao-skill-bank (97 stars, last pushed 5d ago), licensed Apache-2.0. It adds 175 tokens to every session and 4,363 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
data-engineering
A guide to data engineering, the work of collecting, moving, processing, and checking data. It covers workflow tools such as Airflow, Dagster, and Prefect, plus streaming and data-quality tools.
trulens-evaluation-workflow
Systematically evaluate your LLM application with TruLens.
molecular-rag
Retrieve structurally similar compounds with known properties from ChEMBL/ZINC to ground predictions and inform optimization. Based on MolRAG (Xian 2025, ACL).
ai
An index of skills for AI and language-model work, including agent development, security, retrieval-augmented generation, prompting, and evaluation. Retrieval-augmented generation gives a model relevant information from a knowledge base before it answers.
vs-crawler
Crawl websites (news, blogs, papers, GitHub, product docs, RSS feeds) into a fixed-schema JSONL file, then create a dataset and a searchable application in Viking AI Search. Supports one-time crawl and scheduled recurring crawl with automatic incremental sync.
using-model-endpoint
Call a registered model endpoint over its native HTTP API from the endpoint's scoped inference kernel (BASEURL preloaded). Load once a task needs predictions from a registered model endpoint.