Skip to content

docs(*): update-ai-assited-dev-npm-redme#160

Open
radomirchev wants to merge 1 commit intomasterfrom
rmrichev/ai-assisted-dev-npm-readme-update
Open

docs(*): update-ai-assited-dev-npm-redme#160
radomirchev wants to merge 1 commit intomasterfrom
rmrichev/ai-assisted-dev-npm-readme-update

Conversation

@radomirchev
Copy link
Copy Markdown
Contributor

Closes #

Additional information (check all that apply):

  • Bug fix
  • New functionality
  • Documentation
  • Demos
  • CI/CD

Checklist:

  • All relevant tags have been applied to this PR
  • This PR includes general feature table updates in the root README.MD
  • This PR includes CHANGELOG.MD updates for newly added functionality
  • This PR contains breaking changes
  • This PR includes behavioral changes and the feature specification has been updated with them

This package ships with **LLM Agent Skills** — structured knowledge files that teach AI coding assistants (GitHub Copilot, Cursor, Windsurf, Claude, JetBrains AI, etc.) how to work with Ignite UI for React. The skill files are included in the [`igniteui-react`](https://www.npmjs.com/package/igniteui-react) package and also live in the [`skills/`](https://github.com/IgniteUI/igniteui-react/tree/master/skills) directory:


Ignite UI for React provides a complete AI toolchain — **Agent Skills**, the **Ignite UI CLI MCP server**, the **Ignite UI Theming MCP server**, and the **MAKER MCP server** — that grounds AI coding assistants in correct component APIs, import paths, and design tokens.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn't go as far as say it's a complete or even a toolchain really.

You can say the Agent skills and MCP-s "enhance the AI coding [workflow]" for example.

Comment on lines +68 to +70
| Component | What it provides | Requires Ignite UI installed? |
|:----------|:----------------|:-----------------------------:|
| **Agent Skills** | Developer-owned instruction packages: import paths, component patterns, decision flows, project conventions | Yes |
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Skills can be be delivered via ig ai-config without the package, that was the point and we're enhancing that to work even without existing project structure so the last value is also "no" here and TBH just drop the last column entirely - doesn't add useful info.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also is project conversion supposed to be the image to design skill? Because it's a bit vague what that should be

| **Theming MCP server** (`igniteui-theming`) | Design tokens, palette tools, CSS custom property generation, WCAG AA contrast validation | No |
| **MAKER MCP server** (`@igniteui/maker-mcp`) | Multi-agent orchestration: decomposes complex tasks into validated, executable step plans | No |

> **Note:** All AI toolchain commands require Ignite UI CLI 16.0.0 or newer.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You mean 15+, right? :)

Comment on lines +77 to +78
You can start evaluating Ignite UI AI assistance with the MCP servers alone — Ignite UI does not need to be installed in your project. Agent Skills become available once you install Ignite UI packages.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
You can start evaluating Ignite UI AI assistance with the MCP servers alone — Ignite UI does not need to be installed in your project. Agent Skills become available once you install Ignite UI packages.

Run this command from the root of your project:

```bash
npx igniteui-cli ai-config
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If you're giving instructions to install the package at the end and the other examples uses ig, use that consistently:

Suggested change
npx igniteui-cli ai-config
ig ai-config

ig new
```

The CLI prompts for project name and theme, installs dependencies, and configures Agent Skills and MCP servers in a single interactive flow. If the CLI is not installed globally, use `npm install -g igniteui-cli`.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Again don't see how explaining the name prompts and the install of deps is useful information. Just explain you can scaffold a new project with pre-configured AI setup, run ig new and follow the prompts.

The install of the package could be pulled ether before the two sections or added as a note under both snippets


### Agent Skills

This package ships with **Agent Skills** — structured, developer-owned instruction packages that tell AI coding assistants exactly how to use Ignite UI for React. The skill files are included in the [`igniteui-react`](https://www.npmjs.com/package/igniteui-react) package and also live in the [`skills/`](https://github.com/IgniteUI/igniteui-react/tree/master/skills) directory.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I liked the original description paragraph we use across repos better, the "developer-owned instruction packages" says a lot of words, but can't tell what it means TBH.

| **customize-theme** | Customize styling using CSS custom properties, Sass, and the theming system in React |
| **optimize-bundle-size** | Reduce bundle size with granular imports, tree-shaking, and lazy loading |

When a Skill is active in the AI client, the agent follows the Skill instead of relying on general training data — which may reference outdated API signatures or import paths. The Skill package is developer-owned: edit the `SKILL.md` to match your team's conventions, add project-specific patterns, and version the package alongside your codebase.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not sure we need to explicitly explain what that is, could easily just link one of the agent skills mentions to https://agentskills.io/home

Oh, and while true, not sure we want to encourage that since it interferes with update options and "version the package" is wrong since it's not a package, so delete the last part at least, if not the whole section

Suggested change
When a Skill is active in the AI client, the agent follows the Skill instead of relying on general training data — which may reference outdated API signatures or import paths. The Skill package is developer-owned: edit the `SKILL.md` to match your team's conventions, add project-specific patterns, and version the package alongside your codebase.
When a Skill is active in the AI client, the agent follows the Skill instead of relying on general training data — which may reference outdated API signatures or import paths.

Comment on lines +111 to +122
### Supported AI Clients

| Client | Configuration method |
|:-------|:--------------------|
| VS Code with GitHub Copilot | `.vscode/mcp.json` |
| Cursor | `.cursor/mcp.json` |
| Claude Desktop (macOS) | `~/Library/Application Support/Claude/claude_desktop_config.json` |
| Claude Desktop (Windows) | `%APPDATA%\Claude\claude_desktop_config.json` |
| Claude Code | `.mcp.json` or the Claude Code MCP CLI command |
| JetBrains AI Assistant | **Tools → AI Assistant → Model Context Protocol (MCP)** |

Agent Skills are compatible with GitHub Copilot via `.github/copilot-instructions.md`, Cursor via `.cursorrules` or `.cursor/rules/`, Windsurf via `.windsurfrules`, and JetBrains AI Assistant via project-level prompt settings.
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please no.

  • We haven't release the update to pick different MCP configs yet, not sure where that list came from, besides being inaccurate.
  • Skills are not instructions (which we set up as well) or rules
  • There are so many agents/tools and a lot of them support skills and almost all of them work with .agents/skills (except claude, which we cover as well), so you can't put a definitive list.
Suggested change
### Supported AI Clients
| Client | Configuration method |
|:-------|:--------------------|
| VS Code with GitHub Copilot | `.vscode/mcp.json` |
| Cursor | `.cursor/mcp.json` |
| Claude Desktop (macOS) | `~/Library/Application Support/Claude/claude_desktop_config.json` |
| Claude Desktop (Windows) | `%APPDATA%\Claude\claude_desktop_config.json` |
| Claude Code | `.mcp.json` or the Claude Code MCP CLI command |
| JetBrains AI Assistant | **Tools → AI Assistant → Model Context Protocol (MCP)** |
Agent Skills are compatible with GitHub Copilot via `.github/copilot-instructions.md`, Cursor via `.cursorrules` or `.cursor/rules/`, Windsurf via `.windsurfrules`, and JetBrains AI Assistant via project-level prompt settings.


### Agent Skills

This package ships with **Agent Skills** — structured, developer-owned instruction packages that tell AI coding assistants exactly how to use Ignite UI for React. The skill files are included in the [`igniteui-react`](https://www.npmjs.com/package/igniteui-react) package and also live in the [`skills/`](https://github.com/IgniteUI/igniteui-react/tree/master/skills) directory.
Copy link
Copy Markdown
Member

@damyanpetev damyanpetev May 11, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not this package as the description clearly states

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants