Use when writing Bevy (Rust ECS) game code - pin and respect the repo's minor version, plugins/systems/queries/resources, schedules with FixedUpdate and Time, states and run conditions, messages vs observers across versions, required components, testing systems with App::update and a manual time strategy, compile-time and allocation hygiene.
中文简介正在准备中。当前显示原文说明。
Use when writing Godot 4 GDScript - static typing everywhere, scene composition with component nodes, call down/signal up, @onready caching, Resources as data duplicated per instance, autoloads only for services, state machines, pooling, physics in _physics_process, and threaded loading.
中文简介正在准备中。当前显示原文说明。
Use when building a 3D web game with three.js, Babylon.js or PlayCanvas - render loop and delta per library, fixed-step physics with Rapier or Havok, explicit dispose of GPU resources, instancing and draw-call budgets, glTF with Draco/meshopt/KTX2, pixel ratio and resize, WebGPU opt-in, React Three Fiber notes, and headless testing.
中文简介正在准备中。当前显示原文说明。
Use when adding or changing tuning values, items, enemies, abilities or levels - put them in the engine's data assets (ScriptableObject, Resource, DataAsset/DataTable, JSON/RON) with validation and stable IDs, keep runtime state out of shared data, inject services instead of singletons, and test data invariants.
中文简介正在准备中。当前显示原文说明。
Use when writing anything that moves, animates, counts down or simulates - per-frame vs fixed-step callbacks per engine, the accumulator loop with interpolation, delta-scaled smoothing, pause and time scale, and determinism.
中文简介正在准备中。当前显示原文说明。
Use when you are about to hand off a change a player can see - the exact steps to look at it running (web game in the browser at two viewports plus a seeded Playwright smoke run for input and console errors; engine games through a headless frame capture), the checklist, and how to report what could not be observed.
中文简介正在准备中。当前显示原文说明。
Use when touching multiplayer code - authoritative server, client prediction and reconciliation, interpolation, lockstep and rollback, server-side validation of every RPC per engine (Netcode, Unreal replication, Godot multiplayer, Colyseus, Bevy crates), bandwidth and versioned messages.
中文简介正在准备中。当前显示原文说明。
Use when a task touches a Jupyter or marimo notebook, asks to productionise or schedule notebook logic, or when you are tempted to do the work in a notebook — extract logic into tested modules, keep the notebook thin and reproducible, and commit it the way the repo does
中文简介正在准备中。当前显示原文说明。
Use when a task asks for a model, a score, a forecast or a prediction-driven feature — pin down the decision, target, unit, prediction time, metric and baseline before writing any modelling code
中文简介正在准备中。当前显示原文说明。
Use when building or changing features, preprocessing, encoders, resampling or train/validation/test splits — keep every fitted transform inside a Pipeline fitted on training folds only, and build time and group features without seeing the future or the same entity twice
中文简介正在准备中。当前显示原文说明。
Use when writing or changing PyTorch code — model, Dataset/DataLoader, training and evaluation loops, determinism, mixed precision, torch.compile, checkpoints with safetensors, Hugging Face models, ONNX export, and tests that run on CPU
中文简介正在准备中。当前显示原文说明。
Use when data crosses a boundary into your code (file, API, warehouse table, upload, model input or output), when a join or cleaning step can change row counts, or when adding pandera, Great Expectations, Pydantic or dbt checks
中文简介正在准备中。当前显示原文说明。
Use when adding or changing a scheduled data or ML pipeline — Dagster assets, Prefect flows, Airflow DAGs or a cron'd entrypoint — idempotent partitioned stages, retries, backfills, quality gates, and testing the pipeline without the scheduler
中文简介正在准备中。当前显示原文说明。
Use when tools from the jupyter, duckdb, dbt, mlflow, huggingface or postgres MCP servers appear in your tool list, or before reaching for one — what each is good for, what it can damage, and the CLI fallback when it is not connected
中文简介正在准备中。当前显示原文说明。
Use when you are about to hand off any number, table, chart or analysis someone will act on (a metric query, a dbt mart, a report, an experiment readout) — the pre-delivery checklist for data quality, calculation, reasonableness and presentation, and the pitfalls that produce plausible wrong numbers
中文简介正在准备中。当前显示原文说明。
Use when dataframe code is slow or runs out of memory, when data outgrows pandas, or when writing pandas 3, Polars or DuckDB code — vectorisation, dtypes, Copy-on-Write, lazy and out-of-core execution, and measuring before and after
中文简介正在准备中。当前显示原文说明。
Use when designing or analysing an A/B test, computing power or sample size, testing whether a difference is real, or writing any statistical claim — effect sizes with intervals, SRM checks, multiple comparisons and causal caveats
中文简介正在准备中。当前显示原文说明。
Use when training, tuning, comparing or evaluating a model — choosing the cross-validation scheme, baselines, tuning inside CV, thresholds and calibration, per-slice metrics, feature importance and error analysis
中文简介正在准备中。当前显示原文说明。
Use when writing tests for data or ML code — what to test at each layer (transforms, schemas, properties, pipeline smoke, model-quality thresholds, determinism and parity), how to build small synthetic fixtures, and what not to test
中文简介正在准备中。当前显示原文说明。
Use when a model is served online (FastAPI, BentoML) or scored in batch, when changing how a model artifact is loaded or versioned, or when adding drift, quality or prediction monitoring and a rollout plan for a new model
中文简介正在准备中。当前显示原文说明。
Use when you meet a new dataset, table, file or source, before building features or a model on it, or when a number looks wrong — profile grain, keys, nulls, cardinality, time coverage and target balance, and persist what you learn as code
中文简介正在准备中。当前显示原文说明。
Use when starting a Python data or ML project, adding its first tooling, or changing pyproject.toml, the lockfile, the package layout or the lint/type/test configuration — detect the repo's tooling first, then uv, ruff, mypy, pytest and a src layout
中文简介正在准备中。当前显示原文说明。
Use when a change trains a model, reports a metric, or touches seeds, data or model versions, the lockfile, MLflow or DVC — what every run must log, seeding, environment pinning, data versioning and keeping tests off the shared tracking server
中文简介正在准备中。当前显示原文说明。
Use when writing or changing analytical SQL or a dbt model — CTE structure, window functions, grain and join correctness, dbt data tests, unit tests, contracts and incremental models, and how to run them
中文简介正在准备中。当前显示原文说明。