Tao Setup
NVIDIAOfficialCreated by NVIDIA Corporation
- Skill
- Workflow
- Cloud and DevOps
- Apache-2.0
- Medium risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The tao-setup skill provides a one-time session bootstrap for the TAO skill bank. It ensures that individual skills, which are typically installed separately, have access to necessary discovery flows, credential checks, and host preflight validations. This skill acts as a manual alternative to the automatic plugin hook, preparing the environment for multi-skill workflows involving data preparation, training, evaluation, and deployment.
What it does
- Verifies the presence of required TAO skill bank scripts and helpers.
- Performs host preflight checks for Docker and NVIDIA GPU availability.
- Validates the presence of required environment variables for credentials.
- Configures NGC registry authentication for container image pulls.
- Installs the token-efficient execution kit for optimized workflow performance.
- Provides guidance on setting up the TAO environment for consistent cross-skill chaining.
When to use it
- When TAO skills have been installed individually rather than as a bundled plugin.
- Before initiating any multi-skill TAO workflow to ensure environment readiness.
- When troubleshooting missing dependencies or environment configurations for TAO skills.
- To initialize the token-efficient execution kit for repeated application workflows.
Advantages
- Ensures consistent environment configuration across disparate TAO skills.
- Automates critical preflight checks for GPU and container runtime requirements.
- Facilitates secure credential management by checking for variables without exposing values.
- Enables optimized execution modes for reduced context and cost.
Usage examples
- “set up TAO skills”
- The agent initiates the bootstrap process, verifying the TAO skill bank path, checking for Docker and GPU availability, and validating required credentials.
- “prepare TAO environment”
- The agent runs the preflight checks and confirms that the necessary environment variables and container registry access are correctly configured.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/nvidia/tao-setup/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip tao-setup.zip -d .agents/skills
Documentation by NVIDIA Corporation
TAO Setup
One-time session bootstrap for the TAO skill bank. TAO skills are standalone — each model, data, and platform skill carries its own pinned container image and instructions — but multi-skill workflows chain them (data prep, train, evaluate, deploy). This skill provides the session-level pieces that make that chaining work when skills are installed individually: the discovery flow, the credential conventions, and the host preflight.
When the full skill bank is installed as a plugin from this repository, a SessionStart hook injects this guidance automatically and you do not need to run this skill. When skills were installed one-by-one from a skills catalog, run this skill first.
Quick Start
Individual skill installs do not include the shared scripts/ helpers. Obtain
a complete TAO skill bank checkout at the same release ref as the installed
skills, then verify it before using any shared helper. If the checkout is
elsewhere, set TAO_SKILL_BANK_PATH to its root. Reuse a plugin-provided path
when present. If no checkout exists, obtain it with user approval, for example
git clone --branch <installed-release-ref> --single-branch https://github.com/NVIDIA-TAO/tao-skill-bank.git <bank-dir>.
Point TAO_SKILL_BANK_PATH at that checkout and stop if the helper check fails.
export TAO_SKILL_BANK_PATH="${TAO_SKILL_BANK_PATH:-$HOME/tao-skill-bank}"
for item in versions.yaml scripts/resolve_tao_model.py scripts/resolve_tao_image.py scripts/check_tao_launch_preflight.py scripts/redact_secrets.py scripts/tao_job_record.py skills/models skills/platform; do
[ -e "$TAO_SKILL_BANK_PATH/$item" ] || { echo "Incomplete TAO skill bank: missing $item under $TAO_SKILL_BANK_PATH" >&2; exit 1; }
done
set -a; source /path/to/.env; set +a # omit if already exported
# 1. Host preflight — most TAO skills dispatch docker containers on a GPU host.
docker info > /dev/null && echo "OK: docker" || echo "MISSING: docker"
nvidia-smi > /dev/null && echo "OK: GPU" || echo "MISSING: NVIDIA GPU/driver"
# 2. Credential presence check — names only, never print values.
for v in NGC_KEY HF_TOKEN WANDB_API_KEY ACCESS_KEY SECRET_KEY S3_BUCKET_NAME S3_ENDPOINT_URL BREV_API_TOKEN; do
[ -n "${!v:-}" ] && echo "SET: $v" || echo "unset: $v"
done
# 3. NGC registry login (needed for nvcr.io image pulls). Key goes over
# stdin — never as an argv flag, where it lands in the process table.
[ -n "${NGC_KEY:-}" ] && printf '%s' "$NGC_KEY" | docker login nvcr.io -u '$oauthtoken' --password-stdin
# 4. Token-efficient execution kit (recommended): preflight + scaffold
# ~/.tao-kit/kit.env. Non-invasive; see skills/core/tao-token-efficient-execution.
bash "${TAO_SKILL_BANK_PATH}/skills/core/tao-token-efficient-execution/scripts/install.sh"If Docker or the NVIDIA host runtime is missing, use the
tao-setup-nvidia-gpu-host skill — it checks and (with approval) installs
NVIDIA driver 580 or newer, CUDA Toolkit 13.0 or newer, and NVIDIA Container
Toolkit 1.19.0 or newer, and can install Docker itself on Debian/RHEL/SUSE-family
hosts. These are TAO-wide minimums. If the selected model's
references/skill_info.yaml declares runtime_requirements.gpu_host, pass
those model-specific minimums to the host setup skill instead.
Credentials
Load a user-approved env file with set -a; source /path/to/.env; set +a in the
same bash call as the command that consumes the variable. This skill never
creates a credentials file for you; the one credential write here is step 3's
docker login, which stores an nvcr.io token in ~/.docker/config.json.
NGC_KEY— nvcr.io image pulls (most skills)HF_TOKEN— gated HuggingFace weights (several model skills)WANDB_API_KEY— experiment tracking (optional)ACCESS_KEY/SECRET_KEY/S3_BUCKET_NAME/S3_ENDPOINT_URL— S3 I/OBREV_API_TOKEN— Brev platform dispatch
Token-efficient execution (recommended for repeated workflows)
Step 4 of the Quick Start installs the execution kit
(skills/core/tao-token-efficient-execution): compiled card packs run
application workflows as fresh per-stage sessions instead of one long
conversation (measured 74-91% cheaper, 4-5x lower peak context, and the only
mode validated on small execution models). Setup only scaffolds the config;
the routing rule applies at launch time: when a card pack exists for the
requested workflow (skills/applications/<skill>/cards/), prefer its
driver.sh over in-conversation stage execution. The tao-launch-workflow
gate still applies in full — the driver launch is the side-effecting action
it reviews. Packs ship today for tao-run-deft-aoi and tao-run-automl; the
kit's authoring prompt compiles packs for other workflows.
Discovery flow (how TAO skills chain)
Read the task skill. Model skills (
tao-train-*,tao-finetune-*) own network specifics; data skills (tao-generate-*,tao-analyze-*,tao-mine-*, …) own transforms; application skills (tao-run-automl,tao-run-deft-aoi, …) compose model + data + platform into workflows.Read the skill's
references/skill_info.yaml(when present) for the structured contract:container_image(a pinned URI), orbackend_contracts.<backend>.container_imagefor a multi-backend frontend; per-actioncommand,mode,config_format,inputs,outputs, and optionalruntime_requirements.gpu_host. Model runtime requirements override the TAO-wide platform defaults for that workflow.Pick an execution platform and read its skill for mounts, env vars, and resource conventions:
tao-run-on-dockerconventions apply to any localdocker run;tao-run-on-slurm,tao-run-on-kubernetes, andtao-run-on-brevcover managed dispatch;tao-run-on-virtualenvruns a Python script docker-free in a local venv. Externally installed platform skills (e.g. kratos) join as peers — no registration needed. The platforms are equal-class peers — if the user has not chosen, ask; never default silently. Every platform skill implements the same four-verb consumer contract (submit/status/logs/cancel) over its native CLI (docker/kubectl/ssh+sbatch/brev exec) — there is nonvidia-tao-sdk.Construct the spec as nested dicts (
{"train": {"num_epochs": 12}}, never flat dotted keys), confirm with the user, then execute the four verbs:tao-launch-workflowdrives the shared launch gate;scripts/tao_job_record.py openmints the job id and bindsresults_dirbefore launch (record-then-launch); the platform skill runssubmit; then monitor withstatus/logs, mapping native states to the fixed vocabularyPENDING RUNNING COMPLETE ERROR CANCELED UNKNOWN.
Conventions all TAO skills follow
Confirm before side effects.
docker run, job submission, pushes, and file mutations outside the working directory need user confirmation first. Installing a missing Python package prerequisite is the one exception: install it by default and report what was installed.Never ask for credentials in chat and never print credential values or the contents of a credentials file; name the missing variable so the user can export it or add it to an env file you then source.
Container images are pinned per skill. Each skill carries the exact image URI it was validated against; do not swap tags silently. Offer overrides only when the skill documents an override path.
Runtime requirements are layered. Platform skills own the default host requirements and the check/install mechanism. A model may override only the minimum versions it has validated by declaring
runtime_requirements.gpu_hostinreferences/skill_info.yaml; pass those values to the shared host setup check rather than changing the defaults for unrelated models.Execution is SDK-free. Job tracking (
scripts/tao_job_record.py), S3/data staging (tao-data-io, storage tiers A/B/C), and multi-node (the SLURM/K8s templates +scripts/nccl_allreduce_probe.py) are built into the bank — nonvidia-tao-sdk. The one exception is AutoML search (tao-run-automl), which uses thenvidia-tao-automlwheel and its transitive SDK.
Optional: Codex agent identity
For Codex sessions, scripts/install-codex-agents.sh registers the TAO skill
marketplace, installs the plugin, and copies the TAO agent identity to
~/.codex/AGENTS.md so it loads in every session:
bash scripts/install-codex-agents.shFiles
- ATTRIBUTION.mdAdded by AutoGineer897 B
- BENCHMARK.md6.8 KiB
- LICENSELicence file added by AutoGineer11.2 KiB
- SKILL.md9.2 KiB
- config/skillspector-baseline.yaml1.2 KiB
- evals/evals.json951 B
- scripts/install-codex-agents.shScript4.9 KiB
- skill-card.md4.1 KiB
- skill.oms.sig5.0 KiB
LICENSE
Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Declares allowed tools, Needs an API key, Needs a command-line tool, Needs Node.js
The author says: Requires bash and Python 3.10+. Docker plus the NVIDIA container toolkit are needed by most downstream TAO skills but are only checked (not installed) here.
Security
What our scanner noticed· Medium risk
- Large base64-encoded blobs
- References to credential files
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Created by
- NVIDIA Corporation
- Publisher
- NVIDIA (Official)
- Source
- NVIDIA/skills · skills/tao-setup · commit fce3cca
- Licence
- Apache License 2.0 (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE-APACHE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by NVIDIA Corporation or NVIDIA.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download tao-setup.zip