Claude Skills by VectorSpaceLab

"Generate/optimize RDKit 3D conformers, compute alignment/RMSD,

"Use optional RDKit Contrib utilities safely: SA Score, NP Score,

"Use for RDKit data-file discovery, pharmacophore

"Calculate RDKit molecular descriptors, fingerprints, bit-vector

"Use for RDKit molecule creation, file/string I/O, validation,

"Use for RDKit reaction SMARTS/RXN workflows, product sanitization,

"Safely edit, build, and test the RDKit repository itself,

"Use RecBole for recommender-system data preparation, model

"Build, validate, and troubleshoot RecBole Config objects, atomic

"Select RecBole model families and implement or diagnose custom

"Run, evaluate, tune, reload, compare, and troubleshoot RecBole

"Use Microsoft Recommenders to prepare recommendation data, train

"Prepare, validate, split, and transform interaction datasets for

"Compute and troubleshoot Microsoft Recommenders offline metrics

"Choose, instantiate, fit, score, and recommend with Microsoft

"Plan Microsoft Recommenders parameter sweeps, benchmark

"Route and operate rlcode reinforcement-learning examples for

"Guides standard Atari Breakout and Pong DQN/PPO workflows with

"Operate DQN, A2C, and PPO CartPole-v1 training, testing,

"Use and adapt GridWorld dynamic programming, tabular control, Deep

"Use hard-exploration Atari workflows for PPO plus RND,

"Use REINVENT4 for molecular design CLI workflows: sampling,

"Prepare, validate, preprocess, filter, standardize, and

"Configure and validate REINVENT4 enumeration workflows, peptide

"Configure REINVENT4 transfer learning and staged reinforcement

"Configure and troubleshoot REINVENT4 sampling runs for Reinvent,

"Build and troubleshoot REINVENT4 scoring configurations,

"Routes RePaint inpainting, config, output-layout, and

"Routes RePaint config selection, custom dataset and mask layout,

"Explain, render, and tune RePaint jump schedules without running

"Use ResNeSt split-attention CNN models across PyTorch, optional

"Use optional Detectron2 ResNeSt/FPN backbones, config extension,

"Use optional MXNet Gluon ResNeSt models, pretrained parameter

"Use ResNeSt's PyTorch package and Torch Hub models,

"Operate the legacy Rex-Gym quadruped reinforcement-learning

"Use Rex-Gym's deterministic kinematics, Bezier gait, constants,

"Install and operate the legacy Rex-Gym Gym/PyBullet quadruped

"Teaches safe Rex-Gym legacy PPO training setup and packaged policy

"Use and maintain RF-DETR for real-time object detection, instance

"Export and deploy RF-DETR models to ONNX, TensorRT, TFLite,

"Use RF-DETR model variants, checkpoints, prediction outputs, video

"Edit RF-DETR source while following project contribution, test,

"Train, evaluate, configure, and validate RF-DETR datasets with the

"Use RFdiffusion for protein backbone generation workflows,

"Design protein-protein binders with RFdiffusion, including

"Configure RFdiffusion auxiliary guiding potentials, including

"Build RFpeptides macrocyclic peptide monomer and binder

"Design scaffolds around fixed motifs or active sites with

"Diversify an existing RFdiffusion backbone or complex with partial

"Build and validate RFdiffusion scaffold-guided fold-conditioning