agentleFS
chrisgliddonchrisgliddon/bevy-skills/.github/copilot-instructions.md

bevy-skills

Copilot instructions29 starsChanged 11 days agoMITView on GitHub

Copilot does not load Agent Skills directly. This file mirrors the key rules so Copilot users in this repo still benefit. 1. Bevy 0.20 only. Do not suggest older APIs except while explaining a migration. Specifically: - Events use MessageWriter / MessageReader (not the older EventWriter / EventReader; double-check the migration skills when reviewing stale code). - Resources are components; scene serialization uses WorldAsset / WorldAssetRoot and the bevy_world_serialization feature. - Text styling uses FontSource and FontSize components; the default…

The file

We split this copilot-instructions.md into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Bevy 0.20 only: Do not suggest older APIs except while explaining a migration.

    line 7
  2. Very specific:

    Feature collections: When configuring Cargo.toml, pick the right Bevy feature set: 2d, 3d, ui (high-level) or mid-level 2d_api, 3d_api, …

    line 20
  3. Distinctive:

    Read CLAUDE.md. The frontmatter rules, lint script, and tester-crate workflow apply to all agents, not just Claude.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 3,779 Copilot instructions files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
copilot-instructions.md

GitHub Copilot — Bevy 0.20

Copilot does not load Agent Skills directly. This file mirrors the key rules so Copilot users in this repo still benefit.

When writing Bevy code

  1. Bevy 0.20 only. Do not suggest older APIs except while explaining a migration. Specifically:
    • Events use MessageWriter / MessageReader (not the older EventWriter / EventReader; double-check the migration skills when reviewing stale code).
    • Resources are components; scene serialization uses WorldAsset / WorldAssetRoot and the bevy_world_serialization feature.
    • Text styling uses FontSource and FontSize components; the default size is FontSize::Rem(1.) (0.20).
    • Lifecycle observers carry the bundle on the event: On<Add<Player>>, not On<Add, Player> (0.20).
    • Pointer events are flat: On<PointerClick>, not On<Pointer<Click>> (0.20).
    • ui::Interaction and ui::widgets::Button are deprecated; use ui_widgets::Button with Hovered and Has<Pressed> (0.20).
    • Custom shaders are WESL (.wesl, import a::b;, @if(DEF)), not naga_oil #import / #ifdef (0.20).
    • NextState::set_if_different, not set_if_neq (0.20).
    • Required components use #[require(...)] on Component derives.
  2. Feature collections. When configuring Cargo.toml, pick the right Bevy feature set: 2d, 3d, ui (high-level) or mid-level 2d_api, 3d_api, ui_api. See skills/bevy-cargo-features/SKILL.md.
  3. Schedule placement. Frame-driven game/presentation logic goes in Update; fixed-step physics and deterministic simulation go in FixedUpdate. Render-world systems belong in Bevy's render schedules, not Update.
  4. No unwrap() in systems. Bevy systems run every frame. Use let ... else or proper error propagation.
  5. Bridge input into fixed ticks. Translate physical input after InputSystems, queue button transitions, accumulate relative motion, and drain both once on the first fixed tick. Carry held buttons/absolute axes to every tick.
  6. Persist stable domain IDs. Save versioned DTOs, not Entity, dense runtime palette indices, catalog order, asset handles, or reflected ECS layout by default.

When editing skills in this repo

Read CLAUDE.md. The frontmatter rules, lint script, and tester-crate workflow apply to all agents, not just Claude.

Where to look

TopicSkill
ECS basicsskills/bevy-core-concepts/
Rebindable input / gamepads / fixed bridgeskills/bevy-input-actions/
Deterministic and visual testingskills/bevy-testing/
Components & required componentsskills/bevy-ecs-components/
Queries & filtersskills/bevy-ecs-queries/
Systems, sets, run conditionsskills/bevy-ecs-systems/
0.19 → 0.20 breaksskills/bevy-migration-0-19-to-0-20/
0.18 → 0.19 breaksskills/bevy-migration-0-18-to-0-19/
0.17 → 0.18 breaksskills/bevy-migration-0-17-to-0-18/
Assets & custom loadersskills/bevy-assets/, skills/bevy-custom-assets/
Save/load schemas and platform persistenceskills/bevy-save-load/
WASM + WebGPUskills/bevy-wasm-webgpu/
Camerasskills/bevy-cameras/
PBR / materialsskills/bevy-pbr-materials/
Renderer selection / custom passesskills/bevy-rendering/
Diagnostics / tracing / render profilingskills/bevy-diagnostics-profiling/
Physics / Rapier / collision queriesskills/bevy-physics/
Game accessibility / adaptive controllersskills/bevy-a11y/
Audio / sinks / spatial / mixingskills/bevy-audio/
Voxel data, meshing, and production runtimeskills/bevy-voxel-data/, skills/bevy-voxel-pipeline/, skills/bevy-voxel-runtime/

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.