---
title: "MCP Server"
description: "Use the @movk/core MCP server to let AI assistants discover functions, read source, and fetch usage examples through the Model Context Protocol."
canonical_url: "https://core.mhaibaraai.cn/en/docs/getting-started/ai/mcp"
---
# MCP Server

> Use the @movk/core MCP server to let AI assistants discover functions, read source, and fetch usage examples through the Model Context Protocol.

## What is MCP?

MCP (Model Context Protocol) is a standardized protocol that enables AI assistants to access external data sources and tools. @movk/core provides an MCP server so AI tools can directly access function information, source code, and usage examples.

## Available Resources

The @movk/core MCP server exposes the following resources for AI discovery and browsing:

| Resource URI | Description |
| --- | --- |
| `resource://movk-core/all-functions` | Complete list of all available functions with metadata and category information |
| `resource://movk-core/tree-operations` | In-depth documentation for tree structure operations (a core feature) |

Use `@` to access these resources in tools like Claude Code.

## Available Tools

| Tool Name | Description |
| --- | --- |
| `list-functions` | List all @movk/core functions with their categories and basic information |
| `get-function` | Get detailed documentation and usage examples for a specific function; provide `functionName` in camelCase or kebab-case |
| `search-functions` | Search functions by `category`, `subcategory`, or `keyword` |

## Available Prompts

The @movk/core MCP server provides guided prompts for common workflows:

| Prompt Name | Description |
| --- | --- |
| `find-function-for-usecase` | Recommend the most suitable function for a given use case |

Use `/` to access these prompts in tools like Claude Code.

## Configuration

The @movk/core MCP server uses HTTP transport and can be configured in various AI assistants.

### Claude Code

> [!NOTE]
> 
> Make sure Claude Code is installed. See the 
> 
> Anthropic docs
> 
>  for installation instructions.

Add the server via CLI:

```bash
claude mcp add --transport http movk-core https://core.mhaibaraai.cn/mcp
```

### Claude Desktop

1. Open Claude Desktop, go to **Settings** > **Developer**
2. Click **Edit Config** to open the local Claude configuration directory
3. Edit `claude_desktop_config.json`:

```json [claude_desktop_config.json]
{
  "mcpServers": {
    "movk-core": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://core.mhaibaraai.cn/mcp"
      ]
    }
  }
}
```

1. Restart Claude Desktop — the @movk/core MCP server is now registered.

### Cursor

**One-click install**: Click the button below to install the @movk/core MCP server directly in Cursor.

<install-button url="https://core.mhaibaraai.cn/mcp">



</install-button>

**Manual configuration**: Go to **Settings** > **Tools & MCP** to add the @movk/core MCP server, or create/edit `.cursor/mcp.json` in your project root:

```json [.cursor/mcp.json]
{
  "mcpServers": {
    "movk-core": {
      "type": "http",
      "url": "https://core.mhaibaraai.cn/mcp"
    }
  }
}
```

### Visual Studio Code

<install-button url="https://core.mhaibaraai.cn/mcp" ide="vscode">



</install-button>

> [!NOTE]
> 
> Requires the 
> 
> GitHub Copilot
> 
>  and 
> 
> GitHub Copilot Chat
> 
>  extensions.

1. Open the VS Code Command Palette (Ctrl/Cmd + Shift + P)
2. Type **Preferences: Open Workspace Settings (JSON)** and execute
3. Create or edit `mcp.json` in your project's `.vscode` directory:

```json [.vscode/mcp.json]
{
  "servers": {
    "movk-core": {
      "type": "http",
      "url": "https://core.mhaibaraai.cn/mcp"
    }
  }
}
```

### Windsurf

1. Go to **Settings** > **Windsurf Settings** > **Cascade**
2. Click **Manage MCPs** > **View raw config**
3. Add the following configuration:

```json [.codeium/windsurf/mcp_config.json]
{
  "mcpServers": {
    "movk-core": {
      "type": "http",
      "url": "https://core.mhaibaraai.cn/mcp"
    }
  }
}
```

## Usage Examples

Once configured, you can ask your AI assistant questions like:

**Discovery & Queries**

- "List all available @movk/core functions"
- "Get the documentation for the `chunk` function"
- "What parameters does `Tree.fromList` accept?"
- "Find array-related functions"

**Tree Structures (Core Feature)**

- "List all tree structure operation functions"
- "I need to convert tree data to a flat list — which function should I use?"
- "How do I filter a tree by a field while keeping ancestor nodes?"

**Use-Case Based Selection**

- "Which function converts a string to kebab-case?"
- "How do I deep-merge multiple config objects?"
- "Show all URL query parameter handling functions"
- "Get usage examples for `useAppStorage`"

The AI assistant will fetch structured data via MCP tools and provide scenario-specific recommendations. Questions like "which should I use?" or "what's recommended?" will automatically trigger the guided prompt.


## Sitemap

See the full [sitemap](https://core.mhaibaraai.cn/sitemap.md) for all pages.