Revdoku compared with other docs tools
Revdoku publishes API and integration docs from GitHub as docs sites for people and AI agents. This page compares Revdoku with Mintlify, GitBook, ReadMe, Fern and Docusaurus. Each fact about another product links to the official page that states it.
Facts last checked on November 8, 2026, in each product’s official documentation.
The table uses these terms:
- MDX: Markdown with components.
- MCP server: a service where AI agents search and read docs.
llms.txt: a file that lists the pages for AI tools.- On some plans: the product gives the feature only on some of its plans.
- Not listed: we did not find the feature in the official documentation of the product.
| Feature | Revdoku | Mintlify | GitBook | ReadMe | Fern | Docusaurus |
|---|---|---|---|---|---|---|
| Docs source | Two-way sync with a GitHub repository, or with one folder of it. Edits in Revdoku go back as a commit or a pull request. You can also upload files. | Two-way sync with GitHub, GitLab, Bitbucket or Azure Repos. The web editor commits to Git. | Two-way sync with GitHub or GitLab | Two-way sync with GitHub, GitLab or Bitbucket. The repository must be empty when you connect it. | Your repository. The Fern CLI publishes the docs, for example from a GitHub Action. Edits in Fern Editor open a pull request. | Your repository. You build the static site and deploy it, for example with GitHub Actions. |
| Page format | Markdown and MDX. A Mintlify repository works as it is. Components show as plain HTML and do not run code. | Markdown and MDX | Markdown files. MDX is not listed. | MDX on every page | Markdown and MDX | Markdown and MDX |
| API reference from OpenAPI | One reference page for each OpenAPI tag, and an API console where people send real requests | API playground pages that send real requests | OpenAPI blocks with a Test it button | An API reference with a Try It! console | An API reference with an API Explorer that sends real requests | Through community plugins, for example docusaurus-openapi-docs |
| MCP server | Each public docs site has one at /mcp. It is read-only and needs no sign-in. |
Each site has one at /mcp. Sites with sign-in also get /authed/mcp. |
Each published site has one at /~gitbook/mcp |
At /mcp, after you turn it on. Its docs search tools need an add-on. |
At /_mcp/server, when Ask Fern is on. Its one tool answers questions. |
Not listed. A community plugin connects the docs to a third-party MCP service. |
llms.txt and Markdown pages |
llms.txt, llms-full.txt and a Markdown copy of each page, on public sites |
llms.txt, llms-full.txt and a Markdown copy of each page |
llms.txt, llms-full.txt and a Markdown copy of each page |
llms.txt and Markdown pages for AI agents |
llms.txt for the site and for each section, and a Markdown copy of each page. Fern does not make llms-full.txt. |
Through community plugins |
| AI answers on the site | Ask AI on public sites: a panel beside the page. It answers from the docs and names the pages that it used. | An assistant that cites its sources, on some plans | An AI Assistant, on some plans | Ask AI, as an add-on or built in on some plans | Ask Fern, with links to the source pages | Through Algolia DocSearch Ask AI, with your own AI model provider |
| Previews of changes | A 24-hour preview site for each push to another branch. A pull request from the same repository shows the preview of its branch. | A preview for each pull request, on some plans | A preview for each pull request. Only people with a GitBook account can open it. | Branches with a preview, on some plans | Preview links. A GitHub Actions workflow can make one for each pull request. | From your host. The deployment guide says that Netlify and Vercel make a preview for each pull request. |
| Custom domain | Your own domain, or <name>.revdoku.site |
Yes, also at a subpath such as /docs |
Yes, on some plans | Yes | Yes, also at a subpath | Through your host. For GitHub Pages, add a CNAME file (deployment guide). |
| Support requests | Support tickets from Ask AI chats, messages left on the site, page feedback and email. A support agent drafts replies from the docs, and a person sends them. | Intercom and Front widgets, and an assistant contact form that emails your team, on some plans | Not listed. In the Docs Embed, the AI Assistant can call your own tools, for example to open a ticket. | Discussion forums, on some plans. A Zendesk widget lets readers open a ticket. | An Intercom Messenger widget | Not listed. A community plugin adds a third-party help desk. |
| Hosting | Revdoku hosts the site and stores its files in the United States. | Mintlify hosts the site. Self-hosting on some plans. | GitBook hosts the site. Its site renderer is open source. GitBook does not recommend self-hosting it without a specific need. | ReadMe hosts the site. Self-hosting is not listed. | Fern hosts the site. Self-hosting on some plans. | You host the static site. Open source under the MIT license. |
Products change their features often. Read the documentation of each product before you choose a tool.
When to choose which tool
These choices follow from the facts in the table:
- If you want docs from GitHub for people and AI agents, with Ask AI and support tickets, choose Revdoku. A Mintlify repository works as it is (move from Mintlify).
- If your team wants to edit pages together in a web editor that commits to Git, choose Mintlify. It also works with GitLab, Bitbucket and Azure Repos.
- If your writers prefer a block-based editor with two-way sync to GitHub or GitLab, choose GitBook.
- If you want an MCP server that can also send requests to your API, choose ReadMe. ReadMe also has discussion forums for your developers, on some plans.
- If you also want to generate SDKs in several languages from the same API definition, choose Fern.
- If you want an open-source static site generator that you build and host yourself, choose Docusaurus. AI answers come from Algolia DocSearch, and OpenAPI pages and
llms.txtcome from community plugins.
A docs portal with Ask AI
and support tickets
Connect a repository, or describe your product to start. Each docs site has search, Ask AI, an API console, an MCP server and support tickets.
- Publishes from GitHub on every push
- Works with Mintlify repositories
- API console from your OpenAPI file
- MCP server and llms.txt for AI agents
- Ask AI and support tickets
- Analytics for people and AI agents