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-cr-its-mininggit 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-cr-its-mining)<a href="https://agentmods.dev/skills/nvidia-tao/tao-skill-bank/tao-run-deft-cr-its-mining"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-run-deft-cr-its-mining/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-cr-its-mining"><img src="https://agentmods.dev/badge/skills/nvidia-tao/tao-skill-bank/tao-run-deft-cr-its-mining.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
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 →
- high Privilege Escalation · line 229 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
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 | $0.00080 | $0.04421 |
| Opus 5 | $0.00040 | $0.02210 |
| Sonnet 5 | $0.00016 | $0.00884 |
| Haiku 4.5 | $0.00008 | $0.00442 |
Grade A, and why
tao-run-deft-cr-its-mining scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: TAO Run DEFT CR ITS Mining
Prerequisites
Before preflight, follow references/host-prerequisites.md; use its selected DEFT_PYTHON for every bundled helper and stop if the dependency probe fails.
Bundled Resources
Resolve DEFT_SKILL_ROOT to the absolute directory containing this installed SKILL.md. The agent or plugin runtime resolves this path; it is not a user input. Invoke bundled helpers with "$DEFT_PYTHON" "$DEFT_SKILL_ROOT/scripts/<name>.py" .... Never require a skill-bank checkout or change the user's working directory to a repository root.
This workflow invokes tao-finetune-cosmos-embed, tao-analyze-gaps-vlm-bcq, tao-mine-nearest-neighbors, and the selected platform skill by registered skill name. Those skills own their credentials, actions, and bundled assets. Cosmos Reason is workflow-owned: resolve images.tao_toolkit.deft_cosmos_reason, generate TOMLs with this workflow's helpers and configured base templates, and submit the exact train/evaluate commands in references/mining-loop.md through the selected platform. Do not invoke tao-finetune-cosmos-reason for planning, templates, action bundles, commands, or hook resolution.
User Inputs (DEFT Workspace and Workflow Configuration Yaml)
The user should provide a DEFT workspace absolute path that will be used for the entire run. This path should not be /workspace because that conflicts with the Cosmos Reason skills.
The layout of the DEFT workspace should be as follows:
<deft_workspace>/
├── data/
├── hf_cache/
├── model/
├── specs/
└── results/
| Path | Purpose |
|---|---|
<deft_workspace>/data/ |
User-provided KPI and training datasets, including LLaVA annotations and media directories. |
<deft_workspace>/hf_cache/ |
Persistent Hugging Face cache used when workflow steps download or reuse HF models. |
<deft_workspace>/model/ |
Local model/checkpoint inputs for the workflow, including the baseline Cosmos Reason checkpoint and any user-provided Cosmos Embed checkpoint. |
<deft_workspace>/specs/ |
User-provided configuration files, including templates for Cosmos Reason train and evaluate, Cosmos Embed inference, and TAO Data Services mining. |
<deft_workspace>/results/ |
All workflow-generated run artifacts, including state, logs, baseline evaluation, Cosmos Embed outputs, mining outputs, and iteration directories. |
What ships with it
27 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.
- assets/cr_base_evaluate.toml 473 B
- assets/cr_base_train.toml 1.2 KB
- assets/default_cosmos_embed_inference.yaml 800 B
- evals/evals.json 3.3 KB
- references/host-prerequisites.md 1.8 KB
- references/mining-loop.md 26 KB
- scripts/cleanup_cosmos_reason_training.py 5.1 KB runs code
- scripts/compute_bcq_accuracy_metrics.py 6.7 KB runs code
- scripts/cosmos_embed_outputs_to_parquet.py 7.1 KB runs code
- scripts/deft_python.sh 2.0 KB runs code
- scripts/find_cosmos_reason_results.py 856 B runs code
- scripts/initialize_workflow.py 3.4 KB runs code
- scripts/log_stage.py 6.8 KB runs code
- scripts/prepare_cosmos_embed_inference.py 16 KB runs code
- scripts/prepare_cosmos_reason_evaluate.py 6.0 KB runs code
- scripts/prepare_cosmos_reason_train.py 15 KB runs code
- scripts/prepare_gap_analysis_predictions.py 6.1 KB runs code
- scripts/prepare_nearest_neighbor_mining.py 10 KB runs code
- scripts/record_mined_paths.py 2.5 KB runs code
- scripts/restore_docker_mount_permissions.py 4.0 KB runs code
- scripts/resume_position.py 6.6 KB runs code
- scripts/summarize_bcq_accuracy_metrics.py 4.6 KB runs code
- scripts/validate_cosmos_embed_output.py 10 KB runs code
- scripts/verify_workflow_yaml.py 12 KB runs code
- scripts/workflow_common.py 13 KB runs code
- tests/test_cosmos_embed_completion.py 8.1 KB runs code
- tests/test_host_dependencies.py 5.8 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.
- 12d ago First seen · 230 lines · 80 tokens per session scan A 94bec80e6d6e
tao-run-deft-cr-its-mining is a skill published in the GitHub repository NVIDIA-TAO/tao-skill-bank (88 stars, last pushed today), licensed Apache-2.0. It adds 80 tokens to every session and 4,421 once invoked, about $0.0004 per session on Opus 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
fair-esm2
Embed proteins with Meta AI's ESM-2 (fair-esm package). Use this skill when: (1) Extracting per-residue or per-sequence embeddings for downstream ML, (2) Masked-LM likelihood / mutation effect scoring, (3) Contact prediction from a sequence.
evo2
Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring…
openfold3
Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.
boltz
Structure prediction for protein, nucleic-acid, and small-molecule complexes with Boltz-2 (Passaro & Wohlwend et al. 2025, github.com/jwohlwend/boltz). Reach for this skill to validate designed binders against a target, to co-fold a protein with a SMILES or CCD ligand, or to get an open-source AlphaFold3 alternative…
scgpt
Embed and annotate single-cell expression data with scGPT, a foundation model for single-cell biology. Use this skill when: (1) Producing cell embeddings from an AnnData for clustering/integration, (2) Zero-shot or fine-tuned cell-type annotation, (3) Gene-level representation for perturbation/GRN tasks. For…
solublempnn
Inverse-fold a backbone with SolubleMPNN — ProteinMPNN retrained on a soluble-PDB subset (Dauparas et al. 2022) — for sequences biased toward cytosolic expression and reduced aggregation. Reach for this skill when designs from vanilla ProteinMPNN are aggregating or going to inclusion bodies, when redesigning a…