Best agent skills for machine learning
We found 323 agent skills for machine learning in the catalog; these are the 15 that rank highest. The order comes from our rank, which weighs adoption, freshness, documentation and publisher identity. Nobody can pay for a place on this list. ai-pipeline-orchestration, cuml-machine-learning and trl-training lead it. cuml-machine-learning and trl-training are published by the company that makes them. ai-pipeline-orchestration comes from the largest repository here (48,428 stars). 11 of the 15 had activity in the last 30 days. Each entry lists why it made the list and links to its install page.
Updated Oct 26, 2026 · Built from catalog data; how we rank and grade
Top 15 of 323
- 1ai-pipeline-orchestration
devops-skills
Orchestrate AI/ML pipelines for data ingestion, model training, batch inference, and RAG indexing using Prefect, Airflow, or Dagster.
- Largest repository here (48,428 stars)
- Active in the last 30 days
- Open source (MIT)
★ 48.4KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 2cuml-machine-learning
langchain-aiOfficial
Use for GPU-accelerated machine learning on tabular data using NVIDIA cuML.
- Published by the vendor
- Repository has 30.6K stars
- Active in the last 30 days
★ 30.6KLast active Oct 23, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 3trl-training
huggingfaceOfficial
Post-train LLMs with TRL (Transformers Reinforcement Learning) — SFT, DPO, GRPO, KTO, and reward-model training.
- Published by the vendor
- Repository has 11.2K stars
- Active in the last 30 days
★ 11.2KLast active Oct 1, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 4nvflare-convert-pytorch
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Convert existing plain or manual PyTorch training code into an NVFLARE federated job using Client API model exchange, local validation, and job export; use…
- Repository has 3.7K stars
- Active in the last 30 days
- Open source (Apache-2.0)
★ 3.7KLast active Oct 8, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 5nvflare-convert-lightning
NVIDIA FLARE Team <federatedlearning@nvidia.com>
Convert existing PyTorch Lightning training code into an NVFLARE federated job using the Lightning Client API patch, local validation, and job export; use…
- Repository has 3.7K stars
- Active in the last 30 days
- Open source (Apache-2.0)
★ 3.7KLast active Oct 8, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 6codonfm-finetune
NVIDIA BioNeMo <bionemofeedback@nvidia.com>
Fine-tune public CodonFM Encodon checkpoints on labeled coding-sequence or coding-variant data using LoRA, head-only, or full fine-tuning.
- Repository has 3.7K stars
- Active in the last 30 days
- Open source (Apache-2.0)
★ 3.7KLast active Oct 9, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 7cdf-anomalygen
NVIDIA <nvidia@nvidia.com>
Use when running the CDF AnomalyGen pipeline over a defect dataset — mask placement, fine-tuning, synthetic defect-image generation (SDG), evaluation, quality…
- Repository has 3.7K stars
- Active in the last 30 days
- Open source (Apache-2.0)
★ 3.7KLast active Oct 21, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 8nv-segment-ct-finetune
NVIDIA MedTech <noreply@nvidia.com>
Runs standard or fixed-channel softmax finetuning of NV-Segment-CT VISTA3D on CT NIfTI image/label datasets, with optional MONAI-native MLflow tracking and…
- Repository has 3.7K stars
- Open source (Apache-2.0)
- Ships helper scripts
★ 3.7KLast active Sep 15, 2026Claude Code
BFreshness: B–Security: not graded yet - 9nv-generate-mr-brain-finetune
NVIDIA MedTech Team
Used for finetuning NV-Generate-CTMR MR-Brain v1 for T1, T2, FLAIR, SWI, or MRA data from a NIfTI datalist.
- Repository has 3.7K stars
- Open source (Apache-2.0)
- Ships helper scripts
★ 3.7KLast active Sep 15, 2026Claude Code
BFreshness: B–Security: not graded yet - 10depth-estimation
SharpAI
Real-time depth map privacy transforms using Depth Anything v2 (CoreML + PyTorch)
- Repository has 3.1K stars
- Open source (MIT)
- Ships helper scripts
★ 3.1KLast active Sep 17, 2026Claude Code, Claude apps, Codex, Cursor +6
BFreshness: B–Security: not graded yet - 11model-scaffold
Aperivue
Use when you need a runnable PyTorch training repo for a medical-imaging task (segmentation, classification, detection, synthesis, self-supervised, or…
- Active in the last 30 days
- Open source (MIT)
- Ships helper scripts
★ 349Last active Oct 4, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 12archetype-review-base
avelikiy
Shared review framework that every domain reviewer (pci, oracle, gov, edtech, healthcare, mlops, etc.) MUST follow.
- Active in the last 30 days
- Open source (MIT)
★ 105Last active Oct 23, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 13nv-generate-vae-finetune
NVIDIA MedTech Team
Used for finetuning the NV-Generate-CTMR MAISI VAE from CT/MRI NIfTI datalists.
- Repository has 3.7K stars
- Open source (Apache-2.0)
- Ships helper scripts
★ 3.7KLast active Jul 8, 2026Claude Code, Claude apps, Codex, Cursor +6
CFreshness: C–Security: not graded yet - 14gru
Zafer-Liu
使用 GRU 对足量时间序列进行预测（forecast deep learning）
- Repository has 2.6K stars
- Active in the last 30 days
★ 2.6KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet - 15
Installs or verifies AMD Quark and ensures the selected Python environment has an accelerator-matched PyTorch.
- Active in the last 30 days
- Open source (MIT)
★ 465Last active Oct 1, 2026Claude Code, Claude apps, Codex, Cursor +6
AFreshness: A–Security: not graded yet