Tao Setup

What it does

When your agent uses it

• Phrases include set up TAO skills
• TAO session setup
• Prepare TAO environment
• TAO getting started

Example prompts

• “set up TAO skills”
• “TAO session setup”
• “prepare TAO environment”
• “/tao-setup”

Requirements

• Python 3
• A Bash shell
• Docker
• A credential in NGC_KEY
• A credential in WANDB_API_KEY
• Compatibility (from SKILL.md): 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.
• Pre-approved tools (allowed-tools): Read, Bash

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Read the task skill. Model skills (tao-train-*, tao-finetune-*)
2. 2Read the skill's references/skill_info.yaml (when present) for the
3. 3Pick an execution platform and read its skill for mounts, env vars,
4. 4Construct the spec as nested dicts ({"train": {"num_epochs": 12}},

What it can do on your machine

Read from SKILL.md and the folder at commit 0e0d506. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Bash

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/ (Shell), which the agent can run.

  Shell commands in SKILL.md call:

  • docker
  • bash

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use docker, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • NGC_KEY
  • HF_TOKEN
  • WANDB_API_KEY
  • ACCESS_KEY
  • SECRET_KEY
  • BREV_API_TOKEN

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  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.

  From compatibility in the SKILL.md frontmatter.

Context cost

Tao Setup loads about 1.8k tokens when it runs. Until then it costs about 113 tokens; SKILL.md has 711 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~113

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NoteMentions a .env fileSKILL.md:33

  set -a; source /path/to/.env; set +a   # omit if already exported

• NoteMentions a .env fileSKILL.md:59

  d env file with `set -a; source /path/to/.env; set +a` in the

• WarningMentions a credentials file (SSH keys, cloud or package-manager tokens)SKILL.md:62

  n`, which stores an nvcr.io token in `~/.docker/config.json`.

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Bash

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from NVIDIA/skills at commit 0e0d506, republished under its Apache-2.0 licence (© NVIDIA). 711 words, ~1,807 tokens.

name

tao-setup

description

One-time session setup and orchestration map for the TAO skill bank. Run this first when the TAO skills were installed individually (e.g. from a public skills catalog) so the session gets the cross-skill discovery flow, credential checks, and host preflight that the bundled plugin hook would otherwise inject automatically. Trigger phrases include "set up TAO skills", "TAO session setup", "prepare TAO environment", "TAO getting started".

allowed-tools

Read, Bash

compatibility

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.

license

Apache-2.0

metadata.author

NVIDIA Corporation

metadata.version

0.1.0

tags

setup, orchestration, discovery

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

bash scripts/install-codex-agents.sh

© NVIDIA, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (scripts) in skills/tao-setup of NVIDIA/skills.

• SKILL.md
• BENCHMARK.md
• config/skillspector-baseline.yaml
• evals/evals.json
• scripts/install-codex-agents.sh
• skill-card.md
• skill.oms.sig

Open the folder on GitHubat commit 0e0d506