> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tydli.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart Guide

> Deploy your first MCP server in 60 seconds

# Quickstart Guide

## Welcome to Tydli

Tydli transforms **any API documentation** into a fully functional Model Context Protocol (MCP) server in seconds. This allows AI agents like Claude, ChatGPT, and Lovable.dev to directly call your APIs.

**Don't have an OpenAPI spec? No problem!** Our AI can analyze PDFs, HTML docs, Markdown files, or even plain text descriptions and automatically generate everything you need.

## Two Paths to Your MCP Server

### 🚀 Path A: Have API Documentation (Any Format)

**Best for:** PDFs, READMEs, HTML docs, text descriptions

1. **Sign up** for a free Tydli account
2. **Click "Deploy New API"** on your dashboard
3. **Choose "✨ Any Format (AI)"** tab
4. **Upload, paste, or link** your API documentation
5. **AI analyzes** and extracts all endpoints automatically
6. **Review & Deploy** — your MCP server is ready!

→ [Learn more about AI Document Processing](/guides/ai-document-processing)

### 📋 Path B: Have OpenAPI Spec Ready

**Best for:** JSON/YAML OpenAPI files (Swagger 2.0, OpenAPI 3.0/3.1)

1. **Sign up** for a free Tydli account
2. **Click "Deploy New API"** on your dashboard
3. **Paste JSON/YAML**, upload a file, or provide a URL
4. **Configure Authentication** (if your API requires it)
5. **Click "Generate MCP Server"** — done in under 60 seconds!

## Configure Authentication

Whether you used AI or uploaded a spec directly:

* Set up API keys, OAuth 2.0, Basic Auth, or custom headers
* Your credentials are encrypted with AES-256 encryption
* Test credentials before deploying to avoid issues

## Connect to Claude Desktop

After deployment:

1. **Copy your MCP server URL** from the dashboard
2. **Set up OAuth** for secure Claude Desktop integration
3. **Configure Claude** with your server details
4. **Start asking Claude** to use your API!

That's it! You now have a working MCP server that AI agents can interact with.

## What's Next?

* Learn about [Core Concepts](/getting-started/core-concepts) to understand how MCP works
* Follow our [OAuth Integration Guide](/guides/oauth-integration) for Claude Desktop configuration
* Explore [Authentication Methods](/guides/authentication-methods) to secure your API
* Check out [Real-World Use Cases](/guides/use-cases) for inspiration

## Need Help?

* Join our [Discord community](https://discord.gg/tamXgrqxAF) for support
* Check [FAQ](/support/faq) for common questions and answers
* Review our [Troubleshooting Guide](/support/troubleshooting) for detailed help