Interactive Editor

In the API Repository, you can add both API Specs and markdown files, which are organized in the same way. Markdown files use the .md or .mdx extension. To add one, click the plus icon at the top of the file tree and choose the Markdown Document tab.

Beyond standard markdown syntax, we’ve built in dozens of commonly used components like Tabs, Admonitions, Cards, and more to help you build interactive documentation.

The editor has three tabs in the upper right:

Markdown: The editor, with a toolbar for formatting, and a preview next to it that updates while you type. Use Toggle Preview in the toolbar to show or hide the preview.
Diff: Your changes compared with the file before them
Preview: The rendered document

The toolbar’s image button lets you add an image, including uploading one to the repository. Commit your changes from the Changes tab, as for API specs.

Interactive Markdown Editor

Writing in your own editor instead? In a clone of the repository, apigit preview renders the same components on your machine and reloads the page when you save.

Built in components

APIGIT extends standard Markdown with interactive components. Click more in the toolbar to insert one. The menu includes:

Admonitions: Note, Info, Tip, Warning, Danger, Check, Success, Help
Tabs, Expandable, Accordion Group, Card, Card Group, Columns, Code Group, Step
Json Schema, Field, Change, Tooltip, Copy Text, Icon
Container, Image Frame, Image Background
Formula, Video, Embedding and Mermaid Diagrams

For a complete component reference, click Help Docs in the toolbar or visit https://markdown.apigit.com

markdown components

Use a Page With AI Tools

Every page of a published document has a Copy page button at the top right, next to the page title. Click it to copy the page as Markdown; the button reads Copied for a moment. The arrow beside it opens a menu:

Copy page: Copy the page as Markdown for LLMs
Open in ChatGPT and Open in Claude: Start a chat about the page. For a public document the assistant is given the page’s address and reads it itself. A page that needs a sign-in cannot be fetched that way: a short one is put into the chat for you, and a long one is copied so you can paste it (the menu item says which)
Copy MCP server URL, Copy Claude Code command, View as Markdown and Open llms.txt: The addresses an AI agent reads the document at; see Use a published API document with AI agents

An API reference page has the same button beside the API title. There Copy page copies the OpenAPI document with its references resolved, and the menu adds Download OpenAPI file.