Skip to main content
Mintlify is a documentation platform built for the AI era. Your content lives in a Git repository as MDX files, and Mintlify builds and deploys your site automatically when you push a change. But unlike traditional documentation platforms, Mintlify treats your documentation as critical infrastructure—the knowledge layer that powers both human developers and AI agents.

​
Why documentation matters more than ever

As AI models get better, the bottleneck shifts from the model to the context you give it. A smarter model reading stale documentation just produces more confident wrong answers. The teams getting the best results from AI aren’t chasing the latest model release—they’re engineering what goes into the context window for each request. Your documentation is now the primary interface through which AI understands your product. When a developer asks their coding assistant how to authenticate with your API, the assistant queries your documentation. When an AI agent evaluates whether to recommend your product, it’s reading your docs. If your documentation is incomplete, poorly structured, or invisible to AI systems, you lose that evaluation without ever knowing it happened. Mintlify was built to solve this problem. Your documentation becomes AI-native infrastructure that serves both humans browsing your site and AI systems consuming it programmatically.

​
The three parts of a Mintlify project

Your repository is the source of truth for your documentation. It contains an MDX file for every page and a docs.json file that configures your site’s navigation, theme, and settings. You can use your own GitHub or GitLab repository, or let Mintlify create one for you during onboarding. The Mintlify dashboard connects to your repository and lets you manage your site. Use it to monitor deployments, configure settings, manage your team, and edit content directly in the browser. Your site, powered by Mintlify. Mintlify builds your site from your repository and deploys it at a .mintlify.app URL by default. When you’re ready, you can point a custom domain to your site.

​
Editing content

There are two ways to edit your content, and you can switch between them freely.
  • Web editor: Edit and publish pages in your browser. The editor commits changes back to your Git repository automatically.
  • CLI and local editor: Clone your repository, run mint dev to preview your site locally, then push changes to deploy.
Multiple team members can work in either workflow at the same time, using Git branches to manage parallel changes. Anyone who can push to your repository can update your content.

​
AI features

Built-in AI features help people and AI find and understand your content, and help you maintain your content. The assistant lets your users ask questions and get cited answers from your content. The agent helps your team create and maintain content by generating updates from scheduled workflows, pull requests merging in your feature repository, or Slack threads. As AI agents ship code faster than teams can document it, automated workflows keep your documentation current without manual intervention. Your documentation is automatically optimized for AI consumption with machine-readable formats (Markdown export, llms.txt, skill.md), an MCP server for real-time queries, and semantic search that understands intent rather than just matching keywords. See AI-native documentation for an overview of all AI features.

​
Next steps

Quickstart

Deploy your first documentation site in minutes.