# MCP Server for AI Tools

The Freddy documentation site includes a built-in **Model Context Protocol (MCP)** server that allows AI coding assistants to search and read our documentation directly. Instead of copying and pasting from docs, your AI tools can query the full Freddy documentation and API reference in real-time.

## MCP Server URL


```
https://api.aitronos.com/mcp
```

No authentication is required. The documentation is public and freely accessible to all AI tools.

## What AI Tools Can Access

When connected, your AI assistant can:

- **Search all documentation** - Guides, tutorials, and conceptual docs
- **Browse API reference** - All 196+ endpoint definitions, parameters, and response schemas
- **Read OpenAPI schemas** - Data models, request/response formats, and validation rules
- **Find code examples** - Implementation examples across multiple languages


## Setup Instructions

### Claude Desktop

Add to your Claude Desktop configuration file:

**macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
**Windows:** `%APPDATA%\Claude\claude_desktop_config.json`


```json
{
 "mcpServers": {
 "freddy-docs": {
 "url": "https://api.aitronos.com/mcp"
 }
 }
}
```

Restart Claude Desktop after saving. The "Freddy AI Documentation" server will appear in your available tools.

### Claude Code (CLI)

Add the server to your project's `.mcp.json` or global settings:


```json
{
 "mcpServers": {
 "freddy-docs": {
 "type": "url",
 "url": "https://api.aitronos.com/mcp"
 }
 }
}
```

### Cursor

1. Open **Settings** > **MCP Servers**
2. Click **Add new MCP server**
3. Enter the URL: `https://api.aitronos.com/mcp`


Or add to `.cursor/mcp.json` in your project root:


```json
{
 "mcpServers": {
 "freddy-docs": {
 "url": "https://api.aitronos.com/mcp"
 }
 }
}
```

### VS Code (GitHub Copilot)

Add to `.vscode/mcp.json` in your workspace:


```json
{
 "servers": {
 "freddy-docs": {
 "type": "http",
 "url": "https://api.aitronos.com/mcp"
 }
 }
}
```

### Windsurf

Add to your MCP configuration:


```json
{
 "mcpServers": {
 "freddy-docs": {
 "serverUrl": "https://api.aitronos.com/mcp"
 }
 }
}
```

## Example Usage

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

- *"How do I authenticate with the Freddy API?"*
- *"What parameters does the create thread endpoint accept?"*
- *"Show me how to use the Python SDK to send a message"*
- *"What MCP tools are available in Freddy?"*
- *"How do I set up a Streamline automation?"*


Your AI assistant will search the documentation and provide accurate, up-to-date answers based on the actual Freddy docs.

## How It Works

The MCP server is hosted on the Freddy API backend and indexes all documentation content at startup. When an AI tool sends a query:

1. The MCP server searches across all documentation pages and API reference content
2. Relevant pages and sections are identified and ranked
3. The content is returned to the AI tool in a structured format
4. Your AI assistant uses this context to provide accurate answers


The server follows the [Model Context Protocol](https://modelcontextprotocol.io/) specification, which is an open standard for connecting AI models to external data sources.

## Troubleshooting

**Server not responding**

- Verify the URL is exactly `https://api.aitronos.com/mcp`
- Check that the docs site is accessible in your browser at `https://docs.aitronos.com`
- Some corporate firewalls may block MCP connections


**AI tool doesn't show the server**

- Restart the AI tool after adding the configuration
- Verify the JSON configuration is valid (no trailing commas, correct syntax)
- Check the AI tool's logs for connection errors


**Results seem outdated**

- The MCP server content is updated whenever the documentation is deployed
- Clear your AI tool's MCP cache if available


## Related Resources

- [MCP Tools Overview](/docs/documentation/core-concepts/mcp-tools) - How Freddy uses MCP tools internally
- [MCP Protocol Details](/docs/documentation/personal-connectors/mcp) - Deep dive into the MCP protocol
- [Model Context Protocol Specification](https://modelcontextprotocol.io/) - Official MCP standard