
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab"Train, evaluate, and export HumanoidVerse policies with PPO,
"Operate ASRT SpeechRecognition Chinese ASR data, acoustic models,
"Route ASRT acoustic model work for Keras/CTC training, evaluation,
"ASRT data configuration, datalist, dictionary, WAV, and speech
"Route ASRT pinyin sequences through its statistical
"Deploy and integrate ASRT HTTP/gRPC serving clients once trained
"Route Asteroid tasks to the right sub-skill for pretrained
"Build or debug Asteroid custom models, blocks, registries, DSP
"Prepare, inspect, and safely smoke-test Asteroid publishable
"Load Asteroid pretrained checkpoints and separate audio tensors,
"Train and evaluate Asteroid recipes with System, datasets, losses,
"Use Astropy core for astronomy units, coordinates, times, tables,
"Operate Astropy installation checks, optional extras,
"Use Astropy cosmology realizations, FLRW classes, distances, ages,
"Use Astropy modeling, fitting, robust statistics, histograms, time
"Use Astropy Table, QTable, unified I/O, FITS, ASCII/ECSV, VOTable,
"Use Astropy Time and coordinates for timescales, formats, frames,
"Use Astropy units, quantities, equivalencies, custom units,
"Use Astropy visualization and convolution for image normalization,
"Operate Astropy FITS WCS and NDData/CCDData workflows, including
"Use Atomic Agents to build schema-driven agents, integrate tools
"Use AtomicAgent, AgentConfig, BaseIOSchema, ChatHistory,
"Use the Atomic Agents example projects as concrete recipes for
"Use the Atomic Agents MCP connector stack to discover tools,
"Work safely inside the Atomic Agents monorepo: tests, docs,
"Use BaseTool, BaseResource, BasePrompt, Atomic Forge tool
"Guides Attention-Gated Networks PyTorch medical imaging workflows
"Guides Attention-Gated Networks ultrasound classification
"Guides Attention-Gated Networks medical image segmentation
"Use jadore801120 attention-is-all-you-need-pytorch for Transformer
"Prepares and inspects Multi30k and WMT-style BPE data for the
"Instantiate and debug the attention-is-all-you-need-pytorch
"Train Transformer translation models with train.py, safe command
"Run and troubleshoot checkpoint translation with translate.py and
"Use audio-diffusion-pytorch for PyTorch waveform diffusion
"Use diffusion wrapper models conditioned on lower-rate waveforms,
"Route waveform diffusion generator, text-conditioning, sampling,
"Use Augmentor for Pillow-based image augmentation pipelines,
"Use Augmentor generator APIs, array generators, framework
"Mask-safe ground truth augmentation and in-memory grouped array
"Choose and configure Augmentor operations, validation ranges, and
"Directory-backed Augmentor Pipeline workflows for folder scanning,
Operate the Deep Researcher Agent repository for autonomous deep-learning experiments, execution backends, provider/tool dispatch, GPU resources, durable progress, and research-support integrations.
Configure providers and safely operate leader-worker dispatch, text tool calls, repository tools, and literature search.
Launch, configure, steer, and safely stop a PROJECT_BRIEF.md-driven THINK→EXECUTE→REFLECT experiment loop.
Select and operate local, SSH, or Slurm experiment execution with bounded, zero-LLM monitoring, safe paths, truthful terminal outcomes, and actionable recovery.
Detect and select local NVIDIA GPUs, report their resource status, and explain the explicitly opt-in GPU keep-alive without hiding hardware or CUDA limitations.
Preserve truthful experiment context with bounded memory, append-only history, safety signals, gates, and read-only progress inspection.
Operate safe Claude/Codex skill installation and research-support routes for papers, conferences, reports, and Obsidian.
Route Auto-PyTorch tabular and forecasting workflows.