Skills worth trusting.
Agent skills from Anthropic, OpenAI, skills.sh, ClawHub, SkillsMP and other reputable sources. Savant assesses them for security risks and evaluates model-generated test cases using independent scoring. Workspaces import any skill into their own repositories through a reviewed pull request. Every listing on each hub is enumerated; packages are fetched, scanned and evaluated most-popular first as daily capacity allows.
tao-train-optical-inspection
Optical Inspection for defect detection using Siamese networks. Compares image pairs to detect manufacturing defects, anomalies, or quality issues. Use when training, evaluating, exporting, or running inference for a TAO Optical Inspection model on AOI / quality-control data. Trigger phrases include "train optical inspection", "AOI defect detection", "Siamese defect classifier", "PCB / manufacturing inspection".
tao-train-action-recognition
Action recognition from video sequences. Supports RGB, optical flow, and joint (multi-stream) input types for classifying temporal actions in video clips. Use when training, evaluating, exporting, or running inference on a TAO action-recognition model. Trigger phrases include "train action recognition", "video action classification", "RGB + optical flow action model", "TAO ActionRecognition".
tao-train-pose-classification
Pose classification using ST-GCN (Spatial Temporal Graph Convolutional Network). Classifies skeleton sequences into action categories from pose-keypoint data. Use when training, evaluating, exporting, or running inference for a TAO pose-classification model. Trigger phrases include "train pose classification", "skeleton action recognition", "ST-GCN", "keypoint sequence classifier".
tao-analyze-changenet-rca
Performs deep Root Cause Analysis (RCA) on NVIDIA TAO Visual ChangeNet classification experiments with image-evidence-driven investigation. Use when analyzing ChangeNet model failures, investigating poor recall / FAR / PASS-NO_PASS metrics, auditing visual inspection pipeline quality, or running an RCA report for an AOI defect-detection model. Trigger phrases include "RCA on my ChangeNet model", "why is my AOI model failing", "audit ChangeNet predictions", "investigate FAR regressions", "root cause analysis on visual-changenet".
ship-web-games
Package, deploy, and verify a playable Three.js or web game. Use for release builds, asset delivery, private/public deployment, production smoke tests, browser proof, release notes, rollback readiness, and cleanup of temporary QA resources.
tao-analyze-gaps-visual-changenet
Performs gap analysis on NVIDIA TAO VCN Classify (Visual Component Net) experiments by invoking the pinned TAO data-services container directly via `docker run … gap_analysis vcn_aoi …` — picks the optimal decision threshold, ranks per-sample weakness, and emits a top-K weakest parquet expanded per-lighting for downstream augmentation. Use when analyzing VCN classification failures, picking SDA augmentation targets, or auditing PASS/NO_PASS boundary cases.
tao-train-depth-anything-v2
Monocular depth estimation using Metric Depth Anything v2 or Relative Depth Anything architectures. Predicts per-pixel depth from single RGB images. Use when training, evaluating, exporting, or running inference for a TAO monocular depth model. Trigger phrases include "train monocular depth", "DepthAnything v2", "metric depth from single image", "monocular depth estimation".
build-hybrid-game-assets
Plan, create, integrate, or audit a hybrid asset pipeline for a Three.js or web game. Use when choosing among imported meshes, procedural 3D geometry, AI-generated reference art, 2D UI media, sprites, VFX, and performance-ready runtime asset delivery.
experimental-design
Designs experiments and studies BEFORE data is collected — choosing a design, randomizing, blocking, and laying out treatment combinations so results are interpretable. Use whenever someone is planning a study, asks how to assign subjects/samples to groups, mentions randomization, blocking, stratification, controls, factorial or fractional-factorial designs, design of experiments (DOE), screening many factors, response-surface optimization, crossover or repeated-measures or split-plot designs, cluster/group randomization, Latin squares, plate layouts, batch/run-order effects, replication vs. pseudoreplication, or sequential/adaptive/group-sequential designs. Trigger even for informal phrasings like "how should I set up this experiment", "how do I avoid confounding", "what's the best way to test these 6 factors", or "assign these mice to conditions". For computing the sample size or power once the design is chosen, use statistical-power; for analyzing data already collected, use statistical-analysis.
tao-analyze-gaps-vlm-bcq
Extract false-positive and false-negative gaps from VLM binary-classification-question (BCQ, yes/no) predictions. Use when the user asks to "analyze VLM BCQ gaps", "extract VLM false positives and false negatives", or identify failure cases from a predictions JSON for DEFT root-cause analysis on a binary-classification VLM workflow.
tao-train-reid
Person re-identification (ReID). Learns discriminative embeddings to match the same person across different camera views, based on metric learning. Use when training, evaluating, exporting, or running inference for a TAO person re-identification model. Trigger phrases include "train ReID", "person re-identification", "cross-camera person matching", "ReID embeddings", "person re-id".
tao-train-rtdetr
RT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with competitive accuracy and supports distillation and quantization for deployment optimization. Use when training, evaluating, distilling, quantizing, exporting, or running inference for a TAO RT-DETR model. Trigger phrases include "train RT-DETR", "real-time DETR", "low-latency object detection", "RT-DETR distillation / quantization".
promote
Graduate a proven pattern from auto-memory (MEMORY.md) to CLAUDE.md or .claude/rules/ for permanent enforcement. Use when the user runs /si:promote or asks to make a learned behavior permanent.
tao-train-mask2former
Mask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with masked attention for high-quality segmentation results. Use when training, evaluating, exporting, quantizing, or running inference for a TAO Mask2Former model. Trigger phrases include "train Mask2Former", "universal segmentation", "panoptic / instance / semantic segmentation", "masked-attention transformer segmenter".
tao-train-dino
DINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with denoising training, multi-scale features, and optional distillation support. Use when training, evaluating, exporting, distilling, quantizing, or running inference for a TAO DINO detector. Trigger phrases include "train DINO", "DETR object detection", "TAO 2D detection", "DINO with distillation".
sop-creator
Creates detailed Standard Operating Procedures (SOPs) for business processes. Use when user needs SOPs, process documentation, operational guides, workflow documentation, or step-by-step instructions for repeatable business processes.
tao-train-segformer
SegFormer for semantic segmentation. Lightweight transformer-based architecture with hierarchical feature extraction, efficient for real-time segmentation tasks. Use when training, evaluating, exporting, quantizing, or running inference for a TAO SegFormer model. Trigger phrases include "train SegFormer", "semantic segmentation", "lightweight transformer segmenter", "real-time semantic segmentation".
tao-train-nvpanoptix3d
NVPanoptix3D for panoptic 3D scene reconstruction from posed RGB images. Produces 3D panoptic segmentation (semantic, instance, and panoptic masks) with occupancy completion. Built on a VGGT backbone with a Mask2Former-style head and 3D frustum reconstruction. Use when training, evaluating, exporting, or running inference for a TAO NVPanoptix3D model. Trigger phrases include "train NVPanoptix3D", "panoptic 3D reconstruction", "3D scene segmentation", "occupancy completion".
tao-train-nvdinov2
NVDINOv2 for self-supervised visual representation learning. Trains vision transformers via self-distillation (teacher-student) without labels and produces general-purpose visual features. Use when training, exporting, or running inference for a TAO NVDINOv2 backbone. Trigger phrases include "train NVDINOv2", "self-supervised ViT pretraining", "DINOv2 backbone", "visual representation learning".
author-game-levels
Author or revise readable, flat-world Three.js game levels. Use for movement and camera routes, collision and navigation, encounter zones, landmarks, objectives, pickups, motivated lighting, visibility, deterministic level data, or desktop and mobile playthrough verification.
viral-reverse-engineering
Use to reverse-engineer why a piece of content went viral (or overperformed) — yours or someone else's — and extract the repeatable mechanism to apply to your own content. Run when the user says "why did this go viral," "break down this viral post/video," "reverse engineer," "what made this work," or wants to learn from viral content. Sources the observable signal first (intake, transcript, screenshots, top comments, visible stats — an agent usually can't watch a video from a link) and never fabricates what it can't see. Reads brand-profile and audience first, deconstructs the piece layer by layer, isolates the real driver, runs a replicability check, extracts the transferable principle, and applies it to the user's niche via the content skills. Mechanism, never a copy; flags non-replicable virality; visible signals only (no WoopSocial analytics). Single-POST teardown only: for the account-level competitive landscape use competitor-analysis; for riding a live trend use trend-jacking.
tune-enemy-ai
Build, debug, balance, or test combat enemy AI for playable action games. Use for aggro, target selection, navigation, spacing, attack choices, telegraphs, retreats, boss behavior, behavior-state machines, and deterministic AI regression tests.
tao-train-metric-learning-recognition
Metric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for retrieval-based matching (e.g., retail product recognition) using triplet / contrastive losses. Use when training, evaluating, exporting, or running inference for a TAO metric-learning recognition model. Trigger phrases include "train metric learning", "ml-recog", "retrieval embeddings", "triplet loss recognition", "fine-grained matching".
tao-train-centerpose
CenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations for 6-DoF object pose estimation. Use when training, evaluating, exporting, or running inference for a TAO CenterPose model. Trigger phrases include "train CenterPose", "6-DoF object pose", "keypoint estimation", "object pose regression".
tao-generate-referring-expressions
Four-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region descriptions, scene captions, grounded referring expressions, and (optionally) verified expressions via VLM distillation. Use when the user wants to generate referring-expression annotations from images with KITTI labels, build region descriptions, produce grouped grounding phrases tied to bboxes, run a double-check verification pass on grounding expressions, auto-label traffic / scene images for referring datasets, or run the image_referring_expression pipeline. Triggers include 'referring expression', 'region description', 'KITTI labels', 'spatial relationship annotation', 'auto-label image referring expression', 'image_referring_expression'.
tao-validate-dataset-format
Run `tao-daft validate` to check NVIDIA TAO DAFT datasets for structure, schema, and cross-reference errors. Do not use for non-DAFT formats. Use when the user asks to validate a DAFT dataset, check DAFT schema, validate a TAO dataset format, or run `tao-daft validate`.
crypto-ta-analyzer
Run multi-indicator technical analysis on crypto or market OHLCV data. Use for deterministic trend, momentum, volume, and divergence analysis.
tao-train-single-step
Standard single-step train/eval/export workflow for any TAO model. Use when training a TAO model on a dataset without iterative data augmentation, AutoML, or DEFT loops. Trigger phrases include "single train run", "train then evaluate then export", "plain TAO training", "normal training", "no AutoML", "skip the loop". Routes through the per-model SKILL.md for action specifics and through `tao-launch-workflow` for platform/credentials/dataset intake.
migrate
Migrate from Cypress or Selenium to Playwright. Use when user mentions "cypress", "selenium", "migrate tests", "convert tests", "switch to playwright", "move from cypress", or "replace selenium".
pr-improver
Runs an autonomous review-and-fix improvement loop over the current branch's changes until a PR review comes back clean, scoped mechanically to the directories the branch touched. Reviews are performed by an installed PR-review skill (default: pr-review-toolkit's review-pr). Use to fix review findings on a branch before opening or updating a pull request ('clean up this branch', 'fix this PR until review passes', 'run review-and-fix on my changes'). NOT for a one-time review — run the PR-review skill directly.
tao-run-automl-deft-pipeline
Run the canonical NVIDIA AOI three-phase training pipeline — Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA → SDG → mining → plain-train retrain), Phase 3 AutoML refinement on the DEFT-augmented dataset. Use when the user asks to "run the AOI workflow", "fine-tune my PCB AOI model end-to-end", "improve my AOI ChangeNet model", or "AOI workflow with AutoML" request — route here instead of tao-run-deft-aoi directly unless the user explicitly asks for the DEFT loop ONLY (e.g. "run JUST the DEFT loop", "skip AutoML, only DEFT"). Also handles the same three-phase pattern for non-AOI DEFT applications — AutoML baseline then DEFT loop warm-started from AutoML's winning HPs then post-DEFT AutoML refinement on the iteration-augmented dataset. Trigger phrases include "run the AOI workflow", "AOI end-to-end", "AutoML + DEFT", "AutoML then DEFT", "tune hyperparameters then DEFT", "DEFT with AutoML at both ends", "warm-start DEFT", "improve my AOI model".
tao-train-pointpillars
PointPillars for 3D object detection from LiDAR point clouds. Encodes point clouds into a pseudo-image via a pillar-based representation, then applies 2D detection — used in autonomous driving and robotics. Use when training, evaluating, exporting, pruning, retraining, or running inference for a TAO PointPillars model. Trigger phrases include "train PointPillars", "LiDAR 3D detection", "point-cloud object detection", "pillar-based 3D detector".
tao-train-mask-auto-encoder
Masked Auto-Encoder (MAE) for self-supervised pretraining and fine-tuning. Masks random patches and reconstructs them to learn visual representations; supports pretrain and finetune stages. Use when training, evaluating, exporting, or running inference for a TAO MAE backbone. Trigger phrases include "pretrain MAE", "self-supervised vision pretraining", "Masked Autoencoder", "Mask Auto-Encoder", "MAE fine-tune".
tao-run-on-local-docker
Local or remote Docker execution for TAO SDK job containers using a Docker daemon with NVIDIA GPU runtime. Use when running TAO jobs on the current machine, a directly attached Docker host, or a remote GPU box exposed through DOCKER_HOST. Trigger phrases include "run locally", "local Docker", "remote Docker", "use my GPU", "run on my machine", "host Docker daemon".
article-prompts-to-skills
Convert an article, tutorial, or prompt pack into focused reusable AgentSkills, one independent capability per skill, with portable instructions, example prompts, working demos, preview screenshots, validation, gallery updates, and a narrow commit. Use when the user asks to turn an article's prompts, tutorial sections, design patterns, interactions, or workflow ideas into complete skills rather than leaving them as prose.
create-video
Create videos from a text prompt using HeyGen's Video Agent. Use when: (1) Creating a video from a description or idea, (2) Generating explainer, demo, or marketing videos from a prompt, (3) Making a video without specifying exact avatars, voices, or scenes, (4) Quick video prototyping or drafts, (5) One-shot prompt-to-video generation, (6) User says "make me a video" or "create a video about X".
dingtalk-message
钉钉消息发送。当用户提到"钉钉消息"、"发消息"、"发通知"、"群通知"、"群消息"、"Webhook"、"机器人消息"、"机器人发消息"、"工作通知"、"单聊消息"、"群聊消息"、"撤回消息"、"消息已读"、"发送Markdown"、"发卡片消息"、"ActionCard"、"@某人"、"@员工"、"at某人"、"提醒某人"、"dingtalk message"、"send message"、"robot message"、"work notification"时使用此技能。支持:群自定义 Webhook 机器人(文本/Markdown/ActionCard/Link/FeedCard + 加签 + @某人)、企业内部应用机器人单聊和群聊发送、消息撤回、已读查询、工作通知等全部消息类操作。
tao-train-mask-grounding-dino
Mask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for open-set segmentation guided by text prompts. Use when training, evaluating, exporting, quantizing, or running inference for a TAO Mask-Grounding-DINO model. Trigger phrases include "train Mask Grounding DINO", "open-vocabulary segmentation", "text-prompted instance segmentation", "grounded mask DETR".
tao-train-ocdnet
OCDNet for scene text detection. Detects arbitrary-oriented text regions in natural images using a differentiable binarization approach. Use when training, evaluating, exporting, pruning, quantizing, retraining, or running inference for a TAO OCDNet model. Trigger phrases include "train OCDNet", "scene text detection", "arbitrary-oriented text boxes", "differentiable binarization detector".
deepseek-ocr
Expert skill for using DeepSeek-OCR, a vision-language model for optical character recognition with context optical compression supporting documents, PDFs, and images.
frontend-design
Strict frontend design guardrails for building modern, minimal-visual-noise, airy, non-generic app UIs. Use whenever creating or changing frontend UI, React components, shadcn/ui themes, Tailwind styles, app layouts, or first-version product screens. This skill complements vertical/domain skills like CPQ, CRM, and similar app builders: follow the selected domain skill's workflow and layout requirements, then enforce this skill's shared visual rules for brand-aware theming, generous spacing, typography, contrast, roundedness, shadows, surface hierarchy, cardless defaults, no-sidebar defaults, and anti-pattern avoidance.
flows-code-review
Run the technical (code) review step of Flows app certification in a local git app. Loads flows-review-checks for the actual bar, then writes artifacts under reviews/code-review/feedback-round-<N>/. Use when the user asks for a Flows code review, technical review, pre-submit review, app certification code review, or "run flows-code-review". Re-run until 0 open Must Fix items remain before moving on to flows-design-review.
tao-train-foundation-stereo
Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D reconstruction. Use when training, evaluating, exporting, or running inference for a TAO FoundationStereo model. Trigger phrases include "train stereo depth", "FoundationStereo", "stereo disparity estimation", "3D reconstruction from stereo".
api-gateway-backend-security-and-rate-limiting
Secures and rate limits backend services behind Google Cloud API Gateway across OpenAPI 2.0 and OpenAPI 3.0.x API configurations, including MCP-enabled gateways. Use when configuring API key or JWT authentication, locking down Cloud Run or Cloud Run functions backends with a gateway service account, reading authenticated claims from X-Apigateway-Api-Userinfo, setting per-project request quotas, or diagnosing HTTP 429 and HTTP 500 Service Control Quota errors. Do not use for Apigee, GKE Gateway API, Cloud Load Balancing custom domain setup, or LLM model routing.
tao-train-fast-foundation-stereo
Real-time stereo depth estimation using FastFoundationStereo (FFS), the distilled bp2 commercial variant of FoundationStereo. Predicts disparity maps from stereo image pairs with ~10× lower latency than full FoundationStereo. Use when training, evaluating, exporting, or running inference for a TAO FastFoundationStereo (FFS) model. Trigger phrases include "train fast stereo", "real-time stereo disparity", "FastFoundationStereo", "distilled stereo depth".
instagram-growth
Use to grow an Instagram account — the Instagram-specific growth strategy that diagnoses where growth is stuck (reach, follow-conversion, retention, or shares) and orchestrates the other skills to fix it. Run when the user says "grow on Instagram," "Instagram growth," "get more followers/reach on Instagram," "my Instagram isn't growing," "Instagram strategy," or wants more reach/followers on IG. Reads brand-profile, social-strategy, and audience first. Built on the 2026 reality that Instagram is a discovery engine where watch time and sends (DM shares) drive non-follower reach. Diagnoses the broken stage of the growth loop and routes to the right skill (reels-script, carousel-writer, profile-optimization, content-calendar, hashtag-strategy, trend-jacking). Judges via native Instagram Insights — not WoopSocial analytics — and treats follower count as vanity.
build-game-monster-system
Build, integrate, audit, or refactor rigged monsters for Three.js and web action games. Use for monster asset contracts, procedural or imported creature rigs, semantic joints and sockets, hurtboxes and attack volumes, combat animation states, enemy-runtime adapters, LODs, deterministic review fixtures, and validating that every monster follows one shared system.
article-highlights
Editorial Remotion article inserts with mandatory format/background intake, layered highlights, zero neon/glow, and mobile QA.