Claude Skills by VectorSpaceLab

Build Graph Nets blocks and high-level Sonnet model architectures.

Operate on Graph Nets GraphsTuple objects backed by TensorFlow tensors.

"Use Graphify to build, query, export, and integrate code/document

"Install Graphify assistant skills, always-on guidance, and hooks

"Export, merge, and integrate Graphify graph outputs safely without

"Troubleshoot Graphify source-format support, language extractors,

"Builds and updates Graphify graph artifacts from local files,

"Query and navigate existing Graphify graphs with CLI commands,

Routes Graphiti SDK, REST service, and MCP server workflows.

"Guides Graphiti Python SDK workflows for ingest, search, backends,

Guides Graphiti MCP tools, transports, config, and smoke checks.

"Guides Graphiti FastAPI REST service routes, configuration,

"Routes Graphormer users to fairseq training, dataset

"Route Graphormer dataset source selection, custom registration,

"Routes Distributional Graphormer (DiG) catalyst, property-guided,

"Graphormer fairseq-train templates for graph prediction and

"Inspect and extend Graphormer fairseq models, tasks, criterions,

"Guide Graphormer pretrained checkpoint loading, fine-tuning, and

"Use Microsoft GraphRAG from CLI or Python: configure

"Initialize, inspect, and troubleshoot GraphRAG settings,

"Initialize GraphRAG workspaces and run, update, inspect, or

"Build and diagnose GraphRAG lower-level package extensions: custom

"Run, plan, and diagnose GraphRAG prompt tuning for indexing prompt

"Run and diagnose GraphRAG global, local, DRIFT, basic, streaming,

"Use Great Expectations GX Core for data context setup, datasource

"Build, save, and run GX Checkpoints; configure actions,

"Choose and configure Great Expectations Data Contexts, metadata

"Connect GX Core data sources, data assets, batch definitions, and

"Create and maintain Great Expectations expectation classes and

"Create and run GX ValidationDefinitions, pass batch and suite

"Use GroundingDINO for open-vocabulary object detection,

"Pseudo-label image folders with GroundingDINO detections, optional

Evaluate GroundingDINO on COCO-style data and diagnose zero-shot AP results.

"Use GroundingDINO for single-image open-vocabulary detection,

"Build safe GroundingDINO web demos and handoffs to segmentation,

"Use gym-pybullet-drones for PyBullet quadrotor simulation,

"Operate gym-pybullet-drones Betaflight SITL workflows, including

"Operate gym-pybullet-drones headless or GUI control-simulation

"Train, smoke-test, save, load, and play PPO hover policies with

"Use Gymnasium for reinforcement-learning environment loops,

"Choose and troubleshoot Gymnasium built-in environment families,

"Use Gymnasium's single-environment API:

"Design Gymnasium action and observation spaces, validate samples,

"Create, step, wrap, and troubleshoot Gymnasium vectorized

"Apply, inspect, author, and troubleshoot Gymnasium

"Use H2O LLM Studio for no-code and CLI LLM fine-tuning,

"Run and troubleshoot the H2O LLM Studio Wave GUI, app runtime,

"Build, inspect, validate, and round-trip experiment configs and

"Prompt trained experiments locally, tune generation parameters,

"Use for H2O LLM Studio model wrappers, losses, metrics, evaluation