Claude Skills by VectorSpaceLab

"Guides agents configuring CrewAI LLM providers, model strings,

"Guides agents using CrewAI memory, knowledge sources, RAG clients

"Guides agents configuring CrewAI telemetry, tracing, observability

"Guides CrewAI repository contributors through monorepo layout,

"Guides agents using CrewAI tools, crewai_tools exports, custom

"Routes CS230 code-example requests to the correct PyTorch or

"Routes PyTorch SIGNS and NER example requests to the correct

"Routes TensorFlow SIGNS and NER example requests to the correct

"Route CubeStudio MLOps platform deployment, customization, and

"Operate CubeStudio backend lifecycle, configuration overlays,

"Route CubeStudio notebook, registry, image catalog, GPU resource,

"Route CubeStudio dataset, metadata, dimension, SQLLab, and ETL

"Operate CubeStudio deployment workflows safely: Docker Compose

Route CubeStudio pipeline, job-template, Argo, and NNI/HPO workflows.

Route CubeStudio model serving, AIHub, chat, and inference-service workflows.

"cuML operating skill for GPU-accelerated classical ML, cuml.accel,

"Data generation, preprocessing, metrics, model selection, feature

"Route cuML multi-GPU and distributed Dask workflows through

Route cuML source-build, libcuml, C++/CUDA API, and native test workflows.

Direct single-GPU cuML Python estimator workflows on NVIDIA GPUs.

Zero-code-change acceleration for scikit-learn, UMAP, and HDBSCAN.

"Guides CUDA-accelerated cuRobo v2 robotics workflows for robot

"Defines cuRobo robot/world collision scenes, sphere-based

"Solves single, batched, collision-aware, and differentiable

"Plans collision-aware c-space, pose, and grasp motions with

"Configures cuRobo trajectory optimization, rollout costs, and CUDA

"Builds cuRobo GPU TSDF/ESDF maps and sensor-observation pipelines

"Retargets frame and tool-pose sequences to cuRobo robot joint

"Build cuRobo robot models from YAML or URDF, inspect joint and

"Route Custom Diffusion workflows for concept data prep, diffusers

"Validate and interpret CustomConcept101 benchmark artifacts for

Extract, compress, validate, and compose Custom Diffusion checkpoint deltas.

"Prepare instance images, prior-preservation bundles, and concept

"Generate images from Custom Diffusion delta checkpoints for SD

"Guide diffusers Custom Diffusion fine-tuning for single concept,

"Operate CVAT for computer-vision annotation, dataset workflows,

"Implement CVAT local auto-annotation functions and reason about

"Operate CVAT through cvat-cli for profiles, auth, task/project

"Move CVAT data between tasks, projects, dataset formats,

"Deploy and administer self-hosted CVAT with Docker Compose,

"Automate CVAT with the Python SDK, high-level repositories,

Router for CVNets training, evaluation, registry, data, and export workflows.

Convert, benchmark, and profile CVNets models.

Parse, edit, and validate CVNets configs and data layouts.

Choose, inspect, and debug CVNets model families.

Train, finetune, resume, and evaluate CVNets models.

"Route D-FINE object-detection setup, training, architecture

"Inspect and modify D-FINE architecture, registry, YAMLConfig,

Select, validate, and safely modify D-FINE dataset and YAML configs.

"Run D-FINE inference, export, deployment, benchmark,