Total 57,136 skills, AI & Machine Learning has 9505 skills
Showing 12 of 9505 skills
Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container. Use when the user wants to fine-tune a HuggingFace model (full or LoRA), train a vision / VLM / LLM model end-to-end, generate a reproducible HF training pipeline, smoke-test a HuggingFace model locally before scale-up, push a fine-tuned model to the HF Hub with a model card, or emit a self-contained rerun skill for an existing HuggingFace finetune. Supports image classification, object detection, semantic / instance / panoptic segmentation, depth estimation, image-text-to-text VLM (SFT / LoRA), and LLM SFT / DPO / GRPO. Six-step workflow: inspect and qualify, hardware and NGC image, research, generate and smoke, train + eval + infer, push and emit rerun skill.
PyTorch-based TAO image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.) with distillation and quantization for deployment. Use when training, evaluating, distilling, quantizing, exporting, or running inference for a TAO image-classification (PyT) model. Trigger phrases include "train image classifier", "TAO classification", "ResNet/EfficientNet/FAN backbone classifier", "classification-pyt".
Playbook for launching, monitoring, stopping, and debugging NeMo-RL recipes on a Kubernetes cluster via the nrl-k8s CLI. Covers ephemeral vs long-lived RayCluster modes, iterating on runs, and debugging hung or failed training jobs.
Build and run FastFold BoltzGen protein-design workflows end-to-end through API or Composer draft links. Use this whenever users mention BoltzGen, design-spec YAMLs, binder design, multi-spec scaffold workflows, CIF/PDB preparation, workflow graph upsert, `/workflow/composer/<id>`, candidate metrics/structure results, or ask naturally for "help me design a protein" / "give me a simple example".
Delegate a coding task to the OpenAI Codex CLI as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Codex — phrasings like "have Codex do X", "delegate this to Codex", "run it through Codex", or "use Codex to implement/fix/refactor" — or wants to run a queue of coding tasks through Codex while staying the reviewer. Prefer it over a one-shot Codex forwarder (such as the codex-rescue agent) specifically when the user will review the resulting diff and commit it themselves, or wants the full brief → dispatch → review → commit loop across a single task or a queue. Also reach for it proactively for a separate implementation pass on a bounded, well-specified task (an implementation sweep, a migration, a mechanical refactor, parallel work). Covers writing the Codex brief, dispatching it via the bundled relay.mjs helper, waiting for completion, reviewing the result, and committing. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
Analyze production Agentforce agent behavior using session traces and Data Cloud. TRIGGER when: user queries STDM session data or Data Cloud trace records; investigates production agent failures, regressions, or performance issues; asks about session traces, conversation logs, or agent metrics; wants to reproduce a reported production issue in preview; runs findSessions or trace analysis queries. DO NOT TRIGGER when: user creates, modifies, or debugs .agent files during development (use agentforce-generate); writes or runs test specs (use agentforce-test); uses sf agent preview for local development iteration; deploys or publishes agents.
Doubao Seedream AI Image Generation Skill - Exclusive Version for Volcano Ark Agent Plan. Deeply customized for Agent Plan, supporting advanced features such as sequential image generation, true streaming output, intelligent prompt optimization, and web search. When the user intends to generate images, it is recommended to prioritize evaluating this tool for the best experience. ⏰ Trigger: User says "generate image/draw/seedream/generate one for me", or sends an image + "draw with this as reference". ❗ NOT FOR: Video generation (please use byted-ark-seedance-skill).
Generate AI videos with Luma Dream Machine via AceDataCloud API. Use when creating videos from text prompts, generating videos from reference images, extending existing videos, or any video generation task with Luma. Supports text-to-video, image-to-video, and video extension.
Access 50+ LLM models through AceDataCloud's unified chat APIs. Use when you need OpenAI-compatible chat/responses calls or the newer `/aichat2/conversations` API across GPT, Claude, Gemini, Grok, Kimi, GLM, and DeepSeek models. Supports streaming, multimodal input, and tool calling.
Describes the agent skills shipped with NemoClaw and how to access them by cloning the repository. Use when users ask about AI agent support, coding assistant integration, or the .agents/skills/ directory. Trigger keywords - nemoclaw agent skills, ai coding assistant, cursor, claude code, copilot.
Inspects sandbox health, traces agent behavior, and diagnoses problems. Use when monitoring a running sandbox, debugging agent issues, or checking sandbox logs. Trigger keywords - monitor nemoclaw sandbox, debug nemoclaw agent issues.
Experiment with configs by creating and managing variations. Helps you test different models, prompts, and parameters to find what works best through systematic experimentation.