AI
AI-powered development with TallStackUI.
Starting from version 3, TallStackUI ships with AI-ready documentation in its
.ai/
directory, designed to give AI assistants (GitHub Copilot, Claude, Cursor, etc.)
full component knowledge. Each component has a dedicated markdown file with attributes, slots, usage examples, and customization options. In addition,
TallStackUI provides an MCP server hosted on this documentation website, giving AI assistants real-time access to
component documentation without
needing local files.
The following component instruction files are available in the .ai/
directory
of the TallStackUI package. Each file contains comprehensive documentation for AI assistants, including
attributes, slots, usage examples, and soft customization options.
| Component | Instructions |
|---|---|
| Alert | |
| Avatar | |
| Avatar Group | |
| Back to Top | |
| Badge | |
| Banner | |
| Boolean | |
| Breadcrumbs | |
| Card | |
| Carousel | |
| Clipboard | |
| Environment | |
| Errors | |
| Icon | |
| Kbd | |
| Key-Value | |
| Link | |
| Stats | |
| Table | |
| Tooltip | |
| Button | |
| Button Circle | |
| Checkbox | |
| Color Picker | |
| Currency | |
| Date Picker | |
| Error | |
| Hint | |
| Input | |
| Input Select | |
| Label | |
| Number | |
| Password | |
| Pin | |
| Radio | |
| Range | |
| Select Native | |
| Select Styled | |
| Tag | |
| Textarea | |
| Time Picker | |
| Toggle | |
| Upload | |
| Command Palette | |
| Dialog | |
| Dropdown | |
| Dropdown Items | |
| Dropdown Submenu | |
| Loading | |
| Modal | |
| Slide | |
| Toast | |
| Dial | |
| Dial Items | |
| Layout | |
| Layout Header | |
| Sidebar | |
| Sidebar Item | |
| Sidebar Separator | |
| Step | |
| Step Items | |
| Tab | |
| Tab Items | |
| Progress Bar | |
| Progress Circle | |
| Rating | |
| Reaction | |
| Signature | |
| Theme Switch | |
| Floating | |
| Wrapper Input | |
| Wrapper Radio |
The Model Context Protocol (MCP) is an open standard that allows AI tools to access external data sources. TallStackUI provides an MCP server hosted on this documentation website, giving AI assistants real-time access to component documentation without needing local files. The MCP server endpoint is available at:
The MCP server provides the following tools for AI assistants:
| Tool | Description |
|---|---|
list_components
|
List all available components, optionally filtered by category. |
get_component
|
Get full documentation for a specific component. |
search_documentation
|
Full-text search across all component documentation. |
search_customization
|
Search CSS class customization options for components. |
search_classes
|
Search for specific CSS classes across all components. Returns matching blocks with override code snippets for Soft Customization. |
The quickest way to connect the MCP server to Claude Code is using the CLI command:
claude mcp add --transport http tallstackui https://tallstackui.com/mcp/tallstackui
For team sharing, create a
.mcp.json
file in your project root. This file can be
committed to version control so all team members have access:
{ "mcpServers": { "tallstackui": { "type": "http", "url": "https://tallstackui.com/mcp/tallstackui" } }}
For personal use across all projects, add the server to your
~/.claude/settings.json
:
{ "mcp": { "mcpServers": { "tallstackui": { "type": "http", "url": "https://tallstackui.com/mcp/tallstackui" } } }}
After configuration, verify the server is connected:
claude mcp list
To connect the MCP server to Cursor,
create a .cursor/mcp.json
file in your project root:
{ "mcpServers": { "tallstackui": { "type": "http", "url": "https://tallstackui.com/mcp/tallstackui" } }}
You can also create this file at ~/.cursor/mcp.json
for global access across all projects.
The quickest way to connect the MCP server to Gemini CLI is using the CLI command:
gemini mcp add --transport http tallstackui https://tallstackui.com/mcp/tallstackui
Alternatively, add the server to your ~/.gemini/settings.json
:
{ "mcpServers": { "tallstackui": { "httpUrl": "https://tallstackui.com/mcp/tallstackui" } }}