Skip to main content
Move your existing documentation to Mintlify with a coding agent, the Mintlify scraper, exported source files, or a manual migration. You can combine methods, for example scraping a public site for a first draft and then filling in private pages from an export.

​
Choose a migration method

​
Platform guides

The easiest way to migrate is with a coding agent. Run mint migrate --from <platform> to give your agent instructions for your platform. To migrate without an agent, follow a guide for instructions on scraping, exporting, or migrating source files.

Docusaurus

Scrape a public site or migrate source files to preserve private pages, versioned docs, localized content, and custom React components.

ReadMe

Export project files to preserve private content and OpenAPI files or scrape a public site for a faster migration.

GitBook

Export each section with Git Sync for the most complete migration or scrape a public site.

Fern

Migrate MDX, assets, API specifications, and docs.yml configuration from the source repository.

Document360

Export every workspace and language as a project ZIP, including articles, categories, and media files.

Other platforms

Use a coding agent for any platform that mint migrate supports, or follow the manual workflow for a project with mixed content sources.

​
Migrate with a coding agent

The CLI prints migration instructions written for AI coding agents. Run mint migrate to see the workflow and the supported source platforms, or start an agent that follows the instructions in the current directory.
Example migration command
See mint migrate for the supported agents, platforms, and flags.

​
Scrape individual pages

The Docusaurus, ReadMe, and GitBook guides use the scraper’s section command, which converts every page it can reach from the URL you give it. To convert a single page instead, use the page command from your migration directory.
Use page to check how the scraper converts your content before a full run, or to fill in individual pages that a section run missed. For a large site, narrow a section run with --filter rather than scraping everything at once.