
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab"Prepare BEIR training pairs and triplets, build evaluators, select
"Use BentoML to author model-serving Services, build and
"Use BentoML CLI command families and BentoCloud workflows safely:
"Manage BentoML model stores, model references, framework save/load
"Configure BentoML logging, metrics, monitoring/data collection,
"Build BentoML Services into Bentos, author bentofile.yaml or
"Author BentoML services and APIs with the current class-based SDK,
"Serve BentoML services locally over HTTP or gRPC and call
"Use BERT-pytorch for corpus preparation, vocabulary building, and
"Use BERT-pytorch to build, load, and debug vocabularies and
"Use BERT-pytorch to instantiate the model, train on a tiny corpus,
"Route BERTopic topic modeling, embedding, vectorizer, labeling,
"Inspect and visualize trained BERTopic models with topic,
"Select, adapt, and troubleshoot BERTopic embedding backends,
Fine-tune BERTopic topic representations, labels, and multi-aspect outputs.
"Save, load, and share BERTopic models locally or on the Hugging
Build, fit, query, and mutate BERTopic topic models.
"Tunes BERTopic topic-term extraction with c-TF-IDF,
"Route BertViz Transformer attention visualization, neuron-view,
Route BertViz head_view and model_view attention-visualization tasks.
Route BertViz neuron_view query/key neuron inspection workflows.
"Routes BEVFormer camera-only 3D detection workflows, from
"Analyze BEVFormer logs and utility outputs while separating safe
"Validate nuScenes + CAN-bus layout and the generated BEVFormer
"Install and statically inspect the legacy OpenMMLab BEVFormer and
"Compose safe distributed BEVFormer training, evaluation, and FP16
"Route BiRefNet image segmentation, matting, inference,
Route BiRefNet configuration, task selection, and dataset layout checks.
"Route BiRefNet image/video inference, mask saving, foreground
"Route BiRefNet model construction, architecture, backbone,
"Route BiRefNet training launches, resume semantics, evaluation
"Routes Big Sleep text-to-image workflows, CUDA setup checks,
"Guides CUDA-based BigGAN-PyTorch training, checkpoint sampling,
"Prepare and route BigGAN-PyTorch ImageFolder, CIFAR10/CIFAR100,
"Operate BigGAN-PyTorch sampling from compatible checkpoints,
"Converts legacy DeepMind TFHub BigGAN generator modules at 128,
"Train, resume, smoke-test, or troubleshoot BigGAN-PyTorch GAN
"Guide CUDA-enabled BindCraft protein-binder design from target PDB
"Install prerequisites, select BindCraft design and filter presets,
"Inspect BindCraft output folders and CSVs, explain filter outcomes
"Validate a BindCraft target PDB and construct a target JSON with
"Operate the Bindu agent microservice framework: Python bindufy
"Author, run, and troubleshoot Python Bindu agents, A2A task flow,
"Operate Bindu CLI, runtime deployment, source packaging, storage
"Operate Bindu Gateway, recipes, peer auth, A2A planning, Inbox
"Operate Bindu's gRPC core, proto contract, and TypeScript SDK
"Operate Bindu security, identity, private catalog, mTLS, and x402
"Route Biopython computational biology workflows across sequence
"Operate Biopython alignment, search result, BLAST, and phylogeny
"Use Biopython SeqIO and AlignIO for offline sequence/alignment