复制安装命令
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
36 Flutter and Dart skills your coding agent loads by itself, sourced only from official documen...
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
来源文件:README.md
36 Flutter and Dart skills your coding agent loads by itself, sourced only from official documentation.
A skill is a folder with a SKILL.md file. Your agent reads the description, decides a task matches, and pulls in the guidance — you don't paste anything into a rules file or remember to @-mention a doc. Install once, and Firebase Auth guidance shows up when you touch auth, Riverpod guidance when you touch providers.
A comprehensive, (almost) non-opinionated collection: everything here is derived from official Flutter, Dart, Firebase, and package documentation. No personal preferences, no invented conventions.
npx skills add evanca/flutter-ai-rules
That's it. The Skills CLI discovers the packages under skills/ and installs them for supported agents.
Browse before installing, or take just one:
npx skills add evanca/flutter-ai-rules --list
npx skills add evanca/flutter-ai-rules --skill flutter-best-practices
This repo is also a plugin marketplace, so agents that support plugins can install the whole set and keep it updated in place. The manifests live at .claude-plugin/, .codex-plugin/, .cursor-plugin/, .agents/plugins/, and plugin.json — all pointing at the same skills/ directory.
Claude Code — plugin marketplaces:
/plugin marketplace add evanca/flutter-ai-rules
/plugin install flutter-ai-skills@flutter-ai-rules
Skills then load as flutter-ai-skills:bloc, flutter-ai-skills:riverpod, and so on.
Codex — plugins:
codex plugin marketplace add evanca/flutter-ai-rules
codex plugin install flutter-ai-skills@flutter-ai-rules
Cursor — plugins: in Cursor, run /add-plugin, or add the repo as a team marketplace under Dashboard → Settings → Plugins → Import from Repo:
https://github.com/evanca/flutter-ai-rules
Antigravity — plugins:
git clone --depth 1 https://github.com/evanca/flutter-ai-rules.git
agy plugin install ./flutter-ai-rules
Windsurf / Devin has no plugin manifest, but Cascade already scans .agents/skills/ and ~/.agents/skills/, so the manual copy below is the install path. GitHub Copilot doesn't read SKILL.md at all — see Rules and combined sets for what to use instead.
Prefer to do it by hand? Copy or symlink any skill folder into your agent's skills directory — .claude/skills/, .cursor/skills/, .codex/skills/, .agents/skills/ (Antigravity, Codex, Windsurf), .windsurf/skills/. Or vendor the whole set into your project:
git clone --depth 1 https://github.com/evanca/flutter-ai-rules.git temp_repo && mkdir -p .skills && cp -r temp_repo/skills/* .skills && rm -rf temp_repo
With a .skills/ folder you can also reference a skill explicitly when you want it: "Read @.skills/bloc/SKILL.md and create test coverage for the new methods."
| Skill | Loads when you're… |
|---|---|
flutter-best-practices | Writing, reviewing, or planning Flutter code |
effective-dart | Writing Dart, naming things, adding doc comments |
dart-3-updates | Using records, patterns, sealed classes, switch expressions |
flutter-app-architecture | Scaffolding a project or refactoring into layers |
architecture-feature-first | Designing folder structure for a new feature |
flutter-errors | Hitting RenderFlex overflows, unbounded constraints, layout errors |
flutter-use-column-row-first | Building responsive layouts with Row, Column, Expanded, Flexible |
flutter-pre-caching | Preloading fonts, images, animations, or config |
| Skill | Loads when you're… |
|---|---|
bloc | Creating a Cubit or Bloc, modeling state, wiring providers |
riverpod | Setting up providers, combining requests, managing disposal |
provider | Consuming state, optimizing rebuilds, using ProxyProvider |
flutter-change-notifier | Setting up ChangeNotifier models and consuming them |
| Skill | Loads when you're… |
|---|---|
testing | Writing unit, widget, or golden tests; fixing flaky tests |
mockito | Generating mocks, stubbing, verifying interactions |
mocktail | Mocking without codegen, registering fallback values |
patrol-e2e-testing | Writing E2E tests that touch native permissions or dialogs |
code-review | Reviewing a PR, branch, or diff |
| Skill | Loads when you're… |
|---|---|
flutterfire-configure | Adding Firebase to a project, running flutterfire configure |
firebase-auth | Setting up auth, managing auth state, social sign-in |
firebase-cloud-firestore | Designing schemas, CRUD, listeners, pagination |
firebase-database | Syncing real-time data, structuring JSON trees |
firebase-storage | Uploading and downloading files, managing metadata |
firebase-analytics | Logging events, setting user properties |
firebase-crashlytics | Capturing fatal and non-fatal errors |
firebase-messaging | Setting up FCM, handling background messages |
firebase-in-app-messaging | Running in-app campaigns |
firebase-remote-config | Implementing feature flags or A/B tests |
firebase-app-check | Configuring attestation and debug tokens |
firebase-cloud-functions | Calling callable functions, handling errors |
firebase-ai | Generating text or chat with Gemini via firebase_ai |
generate-images-with-firebase-ai | Generating or editing images with a Gemini image model (Nano Banana) |
firebase-data-connect | Writing GraphQL queries against Data Connect |
| Skill | Loads when you're… |
|---|---|
accessibility | Working on a11y, WCAG, screen readers, focus order |
inclusive-design | Handling i18n, global name/address forms, low-end devices |
store-listing-assets | Writing store copy to character limits |
revenuecat-testing | Testing purchases, subscriptions, sandbox flows |
developing-genkit-dart | Building AI agents in Dart with Genkit |
Before skills existed, this repo shipped rule files you pasted into a config, plus pre-merged bundles squeezed under Windsurf's character cap. Both still work and both are still updated, but skills are the recommended path — they load contextually instead of consuming your context window on every request.
Reach for these only if your tool has no skills support, or you want one static file you fully control:
rules/ — six broad foundation files: effective_dart.md, flutter_app_architecture.md, flutter_errors.md, dart_3_updates.md, testing.md, code_review.md. Drop them in your project and reference them by name: "Read @rules/effective_dart.md and follow its conventions." Package-specific guidance (Bloc, Riverpod, Firebase, Mockito…) is skills-only now.combined/ — seven topic bundles, each in a full and an __under_6K variant. Paste one into your global or local rules config and you're done. The trimmed variants stay under 6,000 characters to fit Windsurf's global_rules.md hard limit.GitHub Copilot lives here rather than in the skills section, because it has no SKILL.md support. Three options, in the order Copilot documents them:
combined/ bundle to .github/copilot-instructions.md for repo-wide guidance.rules/ files into .github/instructions/ as <name>.instructions.md, each with applyTo: "**/*.dart" frontmatter, so they only load for Dart files.AGENTS.md, so a bundle pasted there works for Copilot, Codex, and Antigravity at once.Size guidance from each tool's own official documentation, for rule/instruction files and for SKILL.md files (accessed 2026-07-11).
| Tool | Rule file — recommended size | SKILL.md — recommended size |
|---|---|---|
| Claude Code | CLAUDE.md: under 200 lines (soft) | Under 500 lines (soft); description 1,536 chars (hard) |
| Cursor | .mdc rule: under 500 lines (soft) | No numeric limit — "keep focused, move detail to separate files" |
| OpenAI Codex | AGENTS.md: no limit stated | Skill bundle: zip ≤ 50 MB, uncompressed file ≤ 25 MB, ≤ 500 files/version (no per-SKILL.md text limit) |
| Google Antigravity | rule file: 12,000 chars each (hard) | No numeric limit stated |
| Windsurf | global_rules.md: 6,000 chars; .windsurf/rules/*.md: 12,000 chars/file (hard) | No numeric limit — "keeps your context window lean" |
| GitHub Copilot | copilot-instructions.md: ≤ 2 pages (soft, approx.) | Not supported — no repo-level SKILL.md |
Notes:
SKILL.md length recommendation (under 500 lines). Cursor, Windsurf, and Antigravity just say "keep it focused/lean" with no figure; OpenAI documents skill-bundle limits (50 MB zip / 25 MB per file / 500 files) rather than a text length; and GitHub Copilot has no repo-level SKILL.md — it uses copilot-instructions.md plus path-specific *.instructions.md (no size limit stated for the latter).CLAUDE.md in full regardless of length, but notes files over 200 lines "consume more context and reduce adherence"; its auto-memory MEMORY.md loads only the "first 200 lines or 25KB, whichever comes first."docs.windsurf.com → docs.devin.ai) and reference .windsurf/rules and .windsurf/skills.combined/ sets under 6,000 characters to satisfy the strictest hard limit above (Windsurf global_rules.md).Official sources: Claude Code — memory · skills | Cursor — rules · skills | OpenAI Codex — AGENTS.md · skills | Google Antigravity — rules · skills | Windsurf — rules & skills | GitHub Copilot — instructions
Every skill is sourced from official documentation — no personal preferences or subjective interpretations. That's intentional. You're free to alter them to taste, but this repo stays objective by sticking to the source.
One consequence worth knowing: skills can contradict each other, because their sources do. If one package recommends a folder layout and another recommends a different one, you'll see both.
Content is re-fetched from upstream docs on a schedule, so skills track the official guidance as it changes rather than freezing at whatever was true when they were written.
Contributions are welcome:
skills/, or a rule in rules/.Include an official documentation link for anything you add or change. That's the one hard requirement — it's what keeps the repo objective and reviewable. If your source isn't already listed in ATTRIBUTION.md, add it there with its license — and if that license isn't a permissive one, restate the guidance instead of quoting it.
Official documentation these skills are built from:
Flutter — App Architecture · Common Errors · Simple State Management
Dart — Effective Dart · Language tour · Records · Patterns · Pattern types · Branches
State management — Bloc · Riverpod · Provider
Testing — Mockito · Mocktail · Patrol
Firebase — Firebase for Flutter · FlutterFire · Multiple flavors with the FlutterFire CLI
MIT — for this repository's own content: the choice of topics, the trigger descriptions, and the wording and structure of every skill.
The underlying documentation keeps its own terms. Most of it is CC BY 4.0 (Flutter, Dart, Firebase), with code samples under BSD-3 or Apache 2.0, package READMEs under MIT or Apache 2.0, and a few sources — Apple, Google Play, RevenueCat — that aren't openly licensed and are therefore restated as fact rather than copied. ATTRIBUTION.md maps every source to its license and to the skills built on it.
If you redistribute this repo or lift a single skill out of it, keep the source links — that's what the CC BY attribution requirement actually asks for.
name: flutter-pre-caching
description: "Use when preloading fonts, asset/network images, Lottie/Rive animations, local JSON/config, warming initial API data, or optimizing Flutter Web startup."
license: MITPre-caching helps avoid jank, loading flashes, font swaps, and delayed first renders.
The key rule:
Pre-cache only what the user will likely see in the next 1 to 2 screens.
Do not pre-cache the whole app. That can slow startup and waste memory.
Use GoogleFonts.pendingFonts() to load the font variants before showing text.
import 'package:flutter/material.dart';
import 'package:google_fonts/google_fonts.dart';
class ExampleSimple extends StatefulWidget {
const ExampleSimple({super.key});
@override
State<ExampleSimple> createState() => _ExampleSimpleState();
}
class _ExampleSimpleState extends State<ExampleSimple> {
late final Future<List<void>> googleFontsPending;
@override
void initState() {
super.initState();
googleFontsPending = GoogleFonts.pendingFonts([
GoogleFonts.poppins(),
GoogleFonts.montserrat(fontStyle: FontStyle.italic),
]);
}
@override
Widget build(BuildContext context) {
final pushButtonTextStyle = GoogleFonts.poppins(
textStyle: Theme.of(context).textTheme.headlineMedium,
);
final counterTextStyle = GoogleFonts.montserrat(
fontStyle: FontStyle.italic,
textStyle: Theme.of(context).textTheme.displayLarge,
);
return FutureBuilder<List<void>>(
future: googleFontsPending,
builder: (context, snapshot) {
if (snapshot.connectionState != ConnectionState.done) {
return const SizedBox();
}
return Column(
mainAxisAlignment: MainAxisAlignment.center,
children: [
Text(
'You have pushed the button this many times:',
style: pushButtonTextStyle,
),
Text(
'0',
style: counterTextStyle,
),
],
);
},
);
}
}
For production and offline support, prefer bundling critical fonts as assets.
flutter:
fonts:
- family: AppFont
fonts:
- asset: assets/fonts/AppFont-Regular.ttf
- asset: assets/fonts/AppFont-Bold.ttf
weight: 700
Use it in the app theme:
MaterialApp(
theme: ThemeData(
fontFamily: 'AppFont',
),
)
Use precacheImage for images that appear soon.
@override
void didChangeDependencies() {
super.didChangeDependencies();
precacheImage(
const AssetImage('assets/images/header.png'),
context,
);
}
Then use the image normally:
Image.asset('assets/images/header.png')
Use precacheImage with NetworkImage.
@override
void didChangeDependencies() {
super.didChangeDependencies();
precacheImage(
const NetworkImage('https://example.com/image.png'),
context,
);
}
Then use it normally:
Image.network('https://example.com/image.png')
Important:
precacheImage warms Flutter’s in-memory image cache. It does not provide long-term offline caching.
For disk caching, use a package like cached_network_image:
dependencies:
cached_network_image: ^latest
Example:
CachedNetworkImage(
imageUrl: 'https://example.com/image.png',
)
@override
void didChangeDependencies() {
super.didChangeDependencies();
final images = <ImageProvider>[
const AssetImage('assets/images/header.png'),
const AssetImage('assets/images/avatar.png'),
const NetworkImage('https://example.com/banner.png'),
];
for (final image in images) {
precacheImage(image, context);
}
}
precacheImage returns a Future<void>, so you can wait before rendering the real UI.
late Future<void> _preloadImagesFuture;
@override
void didChangeDependencies() {
super.didChangeDependencies();
_preloadImagesFuture = Future.wait([
precacheImage(
const AssetImage('assets/images/header.png'),
context,
),
precacheImage(
const NetworkImage('https://example.com/banner.png'),
context,
),
]);
}
@override
Widget build(BuildContext context) {
return FutureBuilder<void>(
future: _preloadImagesFuture,
builder: (context, snapshot) {
if (snapshot.connectionState != ConnectionState.done) {
return const CircularProgressIndicator();
}
return Column(
children: [
Image.asset('assets/images/header.png'),
Image.network('https://example.com/banner.png'),
],
);
},
);
}
[!NOTE] Use
didChangeDependencies, notinitState, when you needcontextforprecacheImage.
Good candidates:
Bad candidates:
class AppPreloader {
const AppPreloader();
Future<void> preload(BuildContext context) async {
await Future.wait([
_precacheImages(context),
_loadCriticalAssets(),
_warmUpInitialData(),
]);
}
Future<void> _precacheImages(BuildContext context) {
return Future.wait([
precacheImage(
const AssetImage('assets/images/home_hero.png'),
context,
),
precacheImage(
const AssetImage('assets/images/logo.png'),
context,
),
]);
}
Future<void> _loadCriticalAssets() async {
await rootBundle.loadString('assets/config/app_config.json');
}
Future<void> _warmUpInitialData() async {
// Example:
// await userRepository.getCurrentUser();
}
}
Usage:
late Future<void> _preloadFuture;
@override
void didChangeDependencies() {
super.didChangeDependencies();
_preloadFuture = const AppPreloader().preload(context);
}
With FutureBuilder:
@override
Widget build(BuildContext context) {
return FutureBuilder<void>(
future: _preloadFuture,
builder: (context, snapshot) {
if (snapshot.connectionState != ConnectionState.done) {
return const SplashScreen();
}
return const HomeScreen();
},
);
}
Useful for feature flags, local app config, mock data, translations, or onboarding content.
import 'package:flutter/services.dart';
final configJson = await rootBundle.loadString(
'assets/config/app_config.json',
);
Example:
class AppConfigLoader {
Future<String> load() {
return rootBundle.loadString('assets/config/app_config.json');
}
}
This is not pre-caching in the Flutter image-cache sense, but it is often the most useful startup optimization.
final userFuture = userRepository.getCurrentUser();
final dashboardFuture = dashboardRepository.getDashboard();
Example:
class StartupData {
const StartupData({
required this.user,
required this.dashboard,
});
final User user;
final Dashboard dashboard;
}
Future<StartupData> loadStartupData() async {
final results = await Future.wait([
userRepository.getCurrentUser(),
dashboardRepository.getDashboard(),
]);
return StartupData(
user: results[0] as User,
dashboard: results[1] as Dashboard,
);
}
If the animation appears immediately, preload it.
final composition = await AssetLottie(
'assets/animations/success.json',
).load();
Use this only for animations shown early. Avoid preloading many large animations on app start.
Rather than triggering the asset load during a build (which causes jank), preload the RiveFile byte data in advance:
import 'package:flutter/services.dart';
import 'package:rive/rive.dart';
// Preload the RiveFile
final data = await rootBundle.load('assets/animations/character.riv');
final riveFile = RiveFile.import(data);
// Display using direct constructor to avoid initialization delays:
RiveAnimation.direct(riveFile);
This ensures that the animation renders immediately upon widget mount.
Most apps do not need this. It can help if you have:
Basic shape:
class CustomShaderWarmUp extends ShaderWarmUp {
@override
void warmUpOnCanvas(Canvas canvas) {
final paint = Paint();
canvas.drawRect(
const Rect.fromLTWH(0, 0, 100, 100),
paint,
);
}
}
Then set it before runApp:
void main() {
PaintingBinding.instance.shaderWarmUp = CustomShaderWarmUp();
runApp(const MyApp());
}
Use this only when profiling shows shader compilation jank.
On Flutter Web, think less:
Think more:
For critical first-screen images, use both browser preload and Flutter precacheImage.
In web/index.html:
<link
rel="preload"
href="assets/assets/images/home_hero.png"
as="image"
>
In Flutter:
await precacheImage(
const AssetImage('assets/images/home_hero.png'),
context,
);
Why both?
precacheImage warms Flutter's image cache.In web/index.html:
<link
rel="preload"
href="https://example.com/banner.png"
as="image"
>
In Flutter:
await precacheImage(
const NetworkImage('https://example.com/banner.png'),
context,
);
Only do this for images the user is almost guaranteed to see.
Declare the font in pubspec.yaml:
flutter:
fonts:
- family: AppFont
fonts:
- asset: assets/fonts/AppFont-Regular.ttf
- asset: assets/fonts/AppFont-Bold.ttf
weight: 700
Preload the critical font in web/index.html:
<link
rel="preload"
href="assets/assets/fonts/AppFont-Regular.ttf"
as="font"
type="font/ttf"
crossorigin
>
Use it in Flutter:
ThemeData(
fontFamily: 'AppFont',
)
Usually preload:
Avoid preloading every weight and italic variant unless the first screen needs them.
Emojis can trigger font fallback work.
Example:
Text('Welcome 👋 🎉')
On web, this can cause:
Simple and effective:
Text('Welcome')
Then show emojis after the app has loaded:
Text('Welcome 👋')
This is often the best option if emojis are decorative.
Add the font to pubspec.yaml:
flutter:
fonts:
- family: NotoColorEmoji
fonts:
- asset: assets/fonts/NotoColorEmoji.ttf
Preload it in web/index.html:
<link
rel="preload"
href="assets/assets/fonts/NotoColorEmoji.ttf"
as="font"
type="font/ttf"
crossorigin
>
Use it where needed:
const Text(
'👋 🎉 ❤️',
style: TextStyle(
fontFamily: 'NotoColorEmoji',
),
)
Important:
Good cases:
Bad cases:
Rule:
Flutter Web startup can also include renderer assets, for example CanvasKit or Skwasm files.
Usually, you do not preload these from Dart. You configure the renderer through Flutter Web initialization.
Example in web bootstrap:
_flutter.loader.load({
config: {
renderer: 'canvaskit',
},
});
Other configs can include things like:
_flutter.loader.load({
config: {
renderer: 'canvaskit',
canvasKitBaseUrl: '/canvaskit/',
},
});
Most apps should let Flutter manage this unless there is a clear hosting or performance reason to change it.
Usually preload:
Usually avoid:
assets/
images/
logo.png
home_hero.png
onboarding_hero.png
fonts/
AppFont-Regular.ttf
AppFont-Bold.ttf
NotoColorEmoji.ttf
config/
app_config.json
animations/
success.json
character.riv
flutter:
assets:
- assets/images/
- assets/config/
- assets/animations/
fonts:
- family: AppFont
fonts:
- asset: assets/fonts/AppFont-Regular.ttf
- asset: assets/fonts/AppFont-Bold.ttf
weight: 700
- family: NotoColorEmoji
fonts:
- asset: assets/fonts/NotoColorEmoji.ttf
<!-- App fonts -->
<link
rel="preload"
href="assets/assets/fonts/AppFont-Regular.ttf"
as="font"
type="font/ttf"
crossorigin
>
<link
rel="preload"
href="assets/assets/fonts/AppFont-Bold.ttf"
as="font"
type="font/ttf"
crossorigin
>
<!-- Only if emojis are critical to first paint -->
<link
rel="preload"
href="assets/assets/fonts/NotoColorEmoji.ttf"
as="font"
type="font/ttf"
crossorigin
>
<!-- Critical images -->
<link
rel="preload"
href="assets/assets/images/logo.png"
as="image"
>
<link
rel="preload"
href="assets/assets/images/home_hero.png"
as="image"
>
[!NOTE] Flutter web asset URLs often include
assets/assets/...because Flutter serves declared assets under theassets/path, while your asset path also starts withassets/.For example:
- Flutter asset:
assets/images/logo.png- Web URL:
assets/assets/images/logo.png
import 'package:flutter/material.dart';
import 'package:flutter/services.dart';
import 'package:google_fonts/google_fonts.dart';
class AppStartupPreloader {
const AppStartupPreloader();
Future<void> preload(BuildContext context) async {
await Future.wait([
_preloadFonts(),
_preloadImages(context),
_preloadLocalConfig(),
_preloadStartupData(),
]);
}
Future<void> _preloadFonts() {
return GoogleFonts.pendingFonts([
GoogleFonts.inter(),
GoogleFonts.inter(fontWeight: FontWeight.w700),
]);
}
Future<void> _preloadImages(BuildContext context) {
return Future.wait([
precacheImage(
const AssetImage('assets/images/logo.png'),
context,
),
precacheImage(
const AssetImage('assets/images/home_hero.png'),
context,
),
]);
}
Future<void> _preloadLocalConfig() async {
await rootBundle.loadString('assets/config/app_config.json');
}
Future<void> _preloadStartupData() async {
// Start your critical first API calls here.
//
// Example:
// await userRepository.getCurrentUser();
// await dashboardRepository.getDashboard();
}
}
Usage:
class StartupGate extends StatefulWidget {
const StartupGate({super.key});
@override
State<StartupGate> createState() => _StartupGateState();
}
class _StartupGateState extends State<StartupGate> {
Future<void>? _startupFuture;
@override
void didChangeDependencies() {
super.didChangeDependencies();
_startupFuture ??= const AppStartupPreloader().preload(context);
}
@override
Widget build(BuildContext context) {
return FutureBuilder<void>(
future: _startupFuture,
builder: (context, snapshot) {
if (snapshot.connectionState != ConnectionState.done) {
return const SplashScreen();
}
return const HomeScreen();
},
);
}
}
Pre-cache this:
Do not pre-cache this:
For Flutter Web specifically:
precacheImage for Flutter image cache.
评论 (0)
暂无评论,成为第一个评论者吧!