Claude Skills by VectorSpaceLab

"Routes EasyCV dataset layouts, annotation conversion, file I/O,

"Routes EasyCV export, pruning, quantization, TorchAccelerator, and

"Routes EasyCV predictor APIs, batch prediction, feature

"Routes EasyCV config-driven training, evaluation, validation, and

"Use EasyOCR to detect and recognize text in images, configure

"Load EasyOCR custom recognition bundles, validate the file triad,

"Use EasyOCR's DBNet detector, check DBNet import/init, and compile

"Run EasyOCR on images and image-like inputs with the Reader API or

"Operate ECCV2022-RIFE video frame interpolation, benchmark

"Select, prepare, and interpret ECCV2022-RIFE benchmark and

"Use ECCV2022-RIFE image-pair, video, and PNG-sequence

"Operate ECCV2022-RIFE training preflight, Vimeo triplet data

"Route EdgeConnect requests for image inpainting setup, data

"Prepare EdgeConnect images, masks, edges, and flists and validate

"Score EdgeConnect outputs with pixel metrics and validate

"Run EdgeConnect checkpoint-backed testing and inference with

"Configure EdgeConnect training, internal validation sampling,

"Use the einops Python package for readable tensor rearrangement,

"Use optional einops backends, Array API functions, framework

"Use einops.einsum, pack, and unpack for named-axis contractions

"Maintain the einops repository safely: focused tests, backend/env

"Core einops tensor operations for rearrange, reduce, repeat,

"Guide the official Elasticsearch Python client for cluster

"Guide Elasticsearch Python client installation, synchronous and

"Guide elasticsearch.dsl query and search composition, mappings,

"Guide the Elasticsearch Python ES|QL builder, safe parameterized

"Guide Elasticsearch Python bulk indexing, streaming, scanning,

"Use for Electricity Maps contrib parser development, capacity

"Use when updating installed capacity data, adding capacity

"Use when maintaining Electricity Maps zone, exchange, data-center,

"Use when adding, debugging, inspecting, or smoke-testing

Operating skill for the Equinox JAX neural-network and PyTree library.

"Use Equinox serialization, runtime errors, debug tools, pretty

Use Equinox filtered JAX transformations over mixed PyTree models.

Use semi-public equinox.internal helpers for advanced JAX-library workflows.

Use Equinox modules, fields, PyTree filtering, and tree surgery correctly.

"Use Equinox neural-network layers, Sequential models, stateful

"Use the fair-esm repository skill for protein language-model

"Use ESM-IF1 for fixed-backbone protein inverse design, sequence

"Load ESM/ESM-2/MSA Transformer models, tokenize protein inputs,

"Predict protein structures from sequences or FASTA files with

"Score deep mutational scan CSVs with ESM-1v or MSA Transformer

"Use when working with ESPnet speech AI toolkit: installation,

"Maintain, test, review, and contribute to ESPnet source code,

"Configure, dry-run, train, resume, fine-tune, and troubleshoot

"Use ESPnet3 System and Hydra stage workflows, config requirements,

"Use ESPnet pretrained and local inference APIs, model-zoo tags,

"Install, verify, and troubleshoot ESPnet environments, optional

"Create, adapt, validate, and troubleshoot ESPnet2 recipes,

"Use Hugging Face Evaluate to load metrics, comparisons, and