Your agent can do more. Find the skill that makes it happen.
Find your next superpower If you are an agent, refer to our llms.txt for full access.Supports training, evaluation, export, pruning, quantization, retraining, and inference for OCDNet. Its scene text detector uses differentiable binarization to locate text regions at arbitrary orientations.
Supports training, evaluation, export, pruning, quantization, retraining, and inference for OCRNet. Recognizes text from cropped text-region images with CTC or attention-based decoders.
Supports training, evaluation, export, quantization, and inference for OneFormer. Its single architecture uses task-conditioned queries to handle panoptic, instance, and semantic segmentation.
Uses Siamese networks to identify manufacturing defects, anomalies, and quality issues from paired images. Supports training, evaluation, export, and inference on automated optical inspection and quality-control data.
Supports training, evaluation, export, pruning, retraining, and inference for PointPillars. Encodes point clouds as pillar-based pseudo-images for 2D detection, with applications in autonomous driving and robotics.
Uses a Spatial Temporal Graph Convolutional Network to classify skeleton sequences into action categories. Supports model training, evaluation, export, and inference from pose-keypoint data.
Learns discriminative embeddings through metric learning to match the same person across different cameras. Supports training, evaluation, export, and inference for TAO person re-identification models.
Supports training, evaluation, distillation, quantization, export, and inference for RT-DETR. The detection transformer targets real-time inference and includes deployment optimization through distillation and quantization.
Supports training, evaluation, export, quantization, and inference for SegFormer. Its lightweight transformer architecture uses hierarchical feature extraction for efficient semantic segmentation.
Handles a single training, evaluation, and export workflow for any TAO model without iterative augmentation, AutoML, or DEFT. Routes action details to model-specific skills and platform, credentials, and dataset intake to tao-launch-workflow.
Prepare calibrated multi-camera data and train, evaluate, run inference, export, or quantize TAO Sparse4D 3D detection and tracking models. Includes 2D-to-3D geometric distillation with released LTT weights. Use for Sparse4D adaptation and temporal multi-camera perception workflows.
Supports training, evaluation, export, and inference for PCB inspection and automated optical defect detection. Compares image pairs to produce PASS/NO_PASS classifications or change-segmentation masks.
Skills give your agent reusable instructions for a specific job. Pick one, read what it does, and bring it into your workflow.
A name only tells half the story. Search the full description and instructions to find the right fit.
Read the skill, visit its source, and see exactly what you’re adding to your agent.
Copy the install command from a skill page and run it in your project.
npx skillycli add owner/repo --skill name