AI-Assisted Development
AI coding agents like Claude Code, Cursor, and GitHub Copilot can significantly accelerate your development — but only if they understand the bundle’s API. To help them get it right, this repository ships a dedicated agent skill that teaches them exactly how UX DataTables works.
What the Skill Teaches
The skill captures the conventions you would otherwise have to explain to an agent yourself. It covers:
| Topic | What the agent learns |
|---|---|
| Defining a DataTable | Extending `AbstractDataTable`, the `#[AsDataTable(Entity::class)]` attribute, and the `configure*()` hooks |
| Columns | The available column types and shared `AbstractColumn` methods |
| Processing mode | When to use client-side vs server-side rendering |
| Extensions | Buttons, Select, Responsive, ColumnControl, Scroller, KeyTable, ColReorder, FixedColumns |
| Row actions | Edit/delete actions, permissions, and conditional display |
| Integrations | API Platform and Mercure (both opt-in) |
| Gotchas | Common mistakes and their fixes |
How It Is Organized
The skill follows a progressive-disclosure layout: a concise entry point plus reference files the agent reads only when needed.
Entry point
skills/ux-datatables/SKILL.md gives the agent an overview, a decision tree for client-side vs server-side, and a minimal working example.
On-demand references
The skills/ux-datatables/references/ directory holds focused guides the agent pulls in as the task requires:
defining-a-datatable.mdcolumns.mdserver-side.mdextensions.mdactions.mdapi-platform.mdgotchas.md