Index of every page on this site: llms.txt
All skills
a2ui-project avatar

/coding-principles

@6bf4cc2
by A2UIa2ui-project/a2ui18k stars
1,430
Run it yourself

Enforces engineering, architecture, public API, and code quality standards distilled from human reviewer feedback across A2UI pull requests. Use when authoring new code, refactoring, addressing review feedback, or reviewing PR diffs. Don't use for release publishing or spec blueprint generation without code or PR changes.

  • 17 files
  • 75.1 KB
  • Updated 2 weeks ago
  • GitHub
Use this Skill: https://skilld.dev/gh/a2ui-project/a2ui/coding-principles

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈86 tokens for metadata: the name and description. ≈1.9k when used: this file. ≈17k more on demand in 16 files.

Description uses 4.3% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈343 characters, or 4.3%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Skill: coding-principles

Purpose: Core engineering, architecture, and code quality standards established by human reviewers across A2UI pull requests. Used by agents when authoring changes, refactoring code, addressing review feedback, or reviewing pull request diffs to maintain repository health across all languages.


Specification and Blueprint Grounding

Before modifying or implementing code in any language, consult the authoritative specifications and blueprints:


Progressive Discoverability: Two-Level Subskill Routing

Because the A2UI repository spans multiple languages (typescript/, python/, dart/, swift/) and shared specifications (specification/, catalogs/), agents must use a two-level progressive discovery workflow:

  1. Level 1 (Cross-Language Subskills): Activate language-agnostic principles that govern pull request scope, software design, public API boundaries, specification immutability, and changelogs across the entire monorepo.
  2. Level 2 (Language-Specific Routing): Route by the programming languages modified in the diff to load only the subskills and language guides relevant to those languages.

Step 1: Inspect the PR Diff

Review the modified files and categories of changes:

# In an active branch:
git diff main --stat

Identify both:

  • Cross-cutting functional areas impacted (public exports, specifications, catalogs, changelogs, PR structure).
  • Programming languages and directories touched (typescript/, python/, dart/, swift/, samples/, tools/).

Step 2: Activate Cross-Language Subskills (Language-Agnostic)

Evaluate the cross-language subskills below against the diff and read any activated document in full:

1. PR Hygiene, Atomicity, Scope Control & Software Design
  • Skill: pr-hygiene
  • Activate if: Always applicable when authoring, preparing, or reviewing a pull request.
2. Public API Boundaries, Cross-Framework Design & Changelogs
  • Document: subskills/public-api-and-cross-framework-design.md
  • Activate if:
    • Any package entrypoint or facade (index.ts, public-api.ts, package.json, __init__.py, lib/*.dart, or public Swift API), or any CHANGELOG.md is modified.
    • Any new class, function, interface, type, model, or DI token is exported.
    • Renderer APIs, registration helpers, or adapters across @a2ui/angular, @a2ui/lit, @a2ui/react, a2ui_flutter, or A2UISwiftUI are modified.
    • Imports between sibling packages change.
    • Files under specification/ or catalogs/ (schemas, catalog fixtures, or examples) are modified or added.
    • Breaking changes or public API deprecations are introduced.
    • A new source file is added (verify it has a single responsibility rather than acting as a catch-all types.* or utils.* file).
3. Updating Changelogs (CHANGELOG.md)
  • Skill: update-changelog
  • Activate if: Any CHANGELOG.md is modified or a published package's public API or consumer-observable behavior changes.

Step 3: Route by Language for Language-Dependent Principles

Inspect the file extensions and directories in git diff main --stat and follow the matching language route(s) below. Only load routers and guides for languages present in your diff:

Language / Ecosystem Matching Paths & Extensions Progressive Discovery Target
TypeScript & Web (Lit, Angular, React, Web Core, CSS, HTML) *.ts, *.tsx, *.css, typescript/, samples/client/{lit,angular,react}/, tools/ Read the TypeScript & Web router: subskills/typescript/INDEX.md to activate specific TypeScript, reactive lifecycle, CSS theming, accessibility, DOM testing, and AST public API subskills.
Python (Core, Agent SDK, Builders, Evaluators, Scripts) *.py, python/, eval/, samples/agent/, tools/, scripts/ Read the Python router: subskills/python/INDEX.md to activate Python package architecture, facade conventions, typing, and verification subskills.
Swift & SwiftUI (Core, SwiftUI Adapter, iOS Sample) *.swift, swift/, Package.swift Read the Swift & SwiftUI router: subskills/swift/INDEX.md to activate Swift target architecture, coding standards, and Testing subskills.
Dart & Flutter (Core, Agent SDK, Flutter Adapter) *.dart, dart/, samples/client/flutter/, pubspec.yaml Read the Dart & Flutter router: subskills/dart/INDEX.md to activate Dart analyzer rules, package versioning, and changelog subskills.

Step 4: Progressively Load Activated Subskills

Read only the cross-language subskills activated in Step 2 and the language-specific subskills selected through Step 3 to evaluate your code against their rationales, checklists, and examples without loading unrelated language rules into context.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 6bf4cc2. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 weeks ago
metadata
{
  "internal": true
}

README badge

README badge for a2ui-project/a2ui/coding-principles