Claude Skills by VectorSpaceLab

"Route skrl environment construction through the correct framework

"Build and debug public single-agent skrl 2.1.0

"Route skrl 2.1.0 IPPO/MAPPO multi-agent workflows, framework

"Route and compose public single-agent PyTorch agents, models,

"Build and troubleshoot public skrl 2.1.0 Warp DDPG, PPO, and SAC

"Use sktime for time-series forecasting, classification,

"Validate, convert, load, and serialize sktime data containers and

"Use sktime detection, distance, kernel, and alignment APIs for

"Use sktime splitters, metrics, model_evaluation.evaluate, and

"Build and debug sktime-compatible estimators, extension templates,

"Forecast univariate, exogenous, probabilistic, panel, and

"Use sktime for time series classification, regression, clustering,

"Transform time series, extract features, and compose sktime

"Use SkyPilot to run, manage, and scale AI workloads across clouds,

"Operate SkyPilot interactive clusters safely with CLI launch,

"Configure and troubleshoot SkyPilot clouds, Kubernetes, Slurm, SSH

"Launch and debug SkyPilot managed jobs, recovery, queues, logs,

"Modify the SkyPilot repository safely with structure, style,

"Use SkyPilot's Python SDK, async SDK, API server lifecycle

"Configure and operate SkyPilot SkyServe services, replicas,

"Author, validate, troubleshoot, and adapt SkyPilot task and

"Route Skywork-R1V3 local inference, R1V4 API batch testing, and

"Construct and troubleshoot Skywork-R1V3 evaluation workflows for

"Run or adapt local Skywork-R1V3 inference commands with safe

"Draft, inspect, parse, and summarize Skywork R1V4 API batch tests

"Use SMARTS 2.0.1 for multi-agent autonomous-driving simulation,

"Operate SMARTS through the scl Click CLI, build and run scenarios

"Implement SMARTS Agent policies, AgentSpec packages, versioned zoo

"Define, generate, inspect, and troubleshoot SMARTS scenarios from

"Configure SMARTS sensor observations, optional Panda3D rendering,

"Create and operate SMARTS HiWayEnvV1 and parallel Gymnasium

"Build, debug, and operate Hugging Face smolagents agents, tools,

"Build, configure, run, stream, serialize, load, and debug

"Use smolagents command-line entry points, browser automation CLI,

"Configure smolagents CodeAgent execution safely: local imports,

"Select and configure smolagents Model wrappers for Hugging Face

"Use this sub-skill when defining, validating, sharing, loading, or

"Create, run, debug, deploy, and programmatically inspect Snakemake

"Route Snakemake configuration, sample metadata, schema validation,

"Diagnose Snakemake workflows with linting, graph outputs,

"Plan and diagnose Snakemake 9.23.1 software deployment,

"Run, inspect, scale, and troubleshoot Snakemake workflows from the

"Use Snakemake 9.23.1 programmatically with SnakemakeApi,

"Write, refactor, and troubleshoot Snakemake 9.23.1 Snakefiles,

"Use Snips NLU for intent parsing, slot filling, dataset/resource

"Guide Snips NLU CLI command construction for datasets, training,

"Create, inspect, validate, and convert Snips NLU datasets and

"Use SnipsNLUEngine's Python API for fitting datasets, parsing

"Use snnTorch to build spiking neuron models, encode data into

"Build and debug snnTorch neuron, recurrent layer, state lifecycle,