Claude Skills by VectorSpaceLab

"Helps agents select LTX-2 inference pipelines, build safe

"Helps agents choose and troubleshoot LTX-2 performance backends

"Guides LTX Trainer mode selection, config editing, safe launch

"Routes LTX-Video local inference, model/config selection, and

"Guides LTX-Video local CLI and Python inference workflows,

"Guides LTX-Video model YAML selection, config validation, FP8 and

"Guides direct LTX-Video pipeline, scheduler, VAE, transformer,

"Guides agents using Ludwig declarative machine learning configs,

"Guides agents using Ludwig AutoML, init_config, hyperopt search

"Helps agents create, validate, render, and troubleshoot Ludwig

"Guides agents using Ludwig model loading, prediction, evaluation,

"Guides agents serving Ludwig models, building prediction payloads,

"Guides agents running Ludwig train, experiment, check_install,

"Routes Lumina-T2X tasks across text-to-image generation, image

"Routes Lumina text-to-audio and text-to-music demo, config,

"Routes Lumina-T2I, Lumina-Next-T2I, mini, compositional,

"Routes Lumina image-model training, finetuning, data preparation,

"Routes Flag-DiT, Next-DiT, and Next-DiT-MoE ImageNet training and

"Routes visual-anagram generation, view selection, animation, and

"Routes Luminoth object-detection dataset, training, prediction,

"Manage Luminoth checkpoint indexes, remote downloads, local

"Converts and merges Luminoth object-detection datasets, validates

"Runs Luminoth image and video predictions, demo web serving, and

"Configures, runs, and evaluates Luminoth training jobs, including

"Use Lux/Lux API for Pandas-integrated visual exploratory data

"Configure Lux display, plotting, sampling, custom recommendation

"Use Lux's Pandas-integrated DataFrame and Series intent workflow

"Use Lux semantic data types, temporal and geographic handling, ID

"Use Lux optional PostgreSQL SQL executor workflows with

Construct and export Lux Clause, Vis, and VisList visualizations.

"Operate M-flow memory, graph-routed retrieval, ingestion

"Use M-flow public Python APIs and mflow CLI for core memory

"Operate M-flow ingestion, loader selection, memorization stages,

"Operate M-flow retrieval modes, graph-routed Bundle Search,

"Operate M-flow as a FastAPI service, web UI, MCP server, Docker

"Use m2cgen to transpile fitted Python machine-learning models into

Export a fitted Python ML model to native code with m2cgen via API or CLI.

"Repo-specific operating guidance for MaaNTE Neverness to Everness

"Guides MaaNTE Python CustomAction and CustomRecognition

"Guides MaaNTE user-facing daily, city-tycoon, fishing, heist,

"Guides MaaNTE audio, rhythm, MIDI, Tetris, BagelSpam, and

"Guides MaaNTE map teleport, local route navigation, online map

"Guides MaaNTE Pipeline JSON, task option JSON, i18n, SceneManager,

"Use Maestro for vision-language model fine-tuning, dataset

"Validate Maestro dataset layouts, Roboflow identifiers, COCO

"Use Maestro Florence-2 for fine-tuning command/API construction,

"Use Maestro's PaliGemma 2 recipe for JSON extraction, VQA,

"Route Qwen2.5-VL JSON extraction and COCO object-detection

"Mage AI repo skill for platform operations, pipeline authoring,

Mage AI-assisted pipeline generation, documentation, and model configuration.