Separate wiki pages, cloud documents and uploaded files before choosing a route: manual downloads for a few pages, an API workflow for scheduled automation, or a browser extension for a one-time batch export. Verify folder structure, local images and offline readability instead of judging success by file count.
A Feishu knowledge base is usually a mix of wiki pages, cloud documents, uploaded files and nested folders. One page is easy to open. Moving the whole collection is harder because every location has its own permissions and download path.
Feishu’s help documentation says there is no single export for the entire wiki. Pages need to be handled separately, which is why a real migration usually combines several methods.
Identify what the knowledge base contains
Make a simple inventory before exporting:
| Location | Common content | Main difficulty |
|---|---|---|
| Wiki pages | Project docs, policies and long-lived references | Preserving a large page hierarchy |
| Cloud documents | Regular docs, sheets and presentations | Different export entry points |
| Cloud folders | Uploaded PDFs, images and attachments | Files are handled separately |
| External links | Embedded web pages and third-party systems | They may not work offline |
Decide which material is personal, which is owned by the organization, and what you are permitted to retain. Export only content you can access and are authorized to save.
Three practical routes
Route 1: Download documents one by one
Open a document, choose Download as, and select Markdown, Word or PDF. It works fine for a few files. At dozens or hundreds of pages, the repeated operations become slow and the hierarchy still needs cleanup.
Use this route for fewer than about twenty documents.
Route 2: Open Platform API plus a command-line tool
These tools usually require a custom app, read permissions and a terminal command. They are a good fit for scheduled backups and very large collections.
They also require administrator approval, key management and command-line knowledge. For a one-time personal migration, the setup can cost more than the export itself.
Route 3: A browser extension using the current signed-in session
An extension can read the Feishu pages that are already visible in your browser. It does not require a custom app or a public sharing link.
Its boundary is equally important: it can export only pages you can open and is limited to the browser version of Feishu. It cannot bypass permissions or read content that exists only in the desktop client.
What a usable batch export must preserve
Folder hierarchy
The output should still resemble the original wiki structure. A flat pile of files is difficult to navigate even when every document exported successfully.
Local images
Markdown that keeps remote image URLs is not a complete archive. Images should be saved locally and referenced with relative paths.
Working links
Internal Feishu links, attachments and external URLs need separate checks. Some links stop working after an account leaves the organization, so important material should have a local copy.
Offline readability
Move the output folder to another computer, disconnect from the network, and open several documents. Images, tables and code blocks must still render.
Export with a browser extension
- Open the Feishu wiki in a browser and confirm that every target page is readable.
- Start from the wiki root and select the required range.
- Add multiple documents to the export queue.
- Choose an output location and include images.
- Inspect folders, images and document formatting after export.
Manual export remains practical for small collections. A batch queue becomes useful in the tens-to-hundreds range. Feishu Toolkit Pro provides that queue, and each paid tool includes three free trials.
When an extension is not the right tool
If the job requires unattended scheduled backups, server uploads or integration with an existing automation system, an Open Platform API and command-line workflow will be more appropriate.
If the organization prohibits browser extensions from reading document pages, do not bypass the policy. Confirm the internal rule and choose a compliant route.
The archive is finished only when the folders make sense, documents open, images still resolve and you have permission to retain the content. A large export that fails one of those checks is not a migration.