Welcome!
-
-
- Learn Home
- Core Concepts
- Companies and Departments
- Courses
- Course Status and Visibility
- Modules
- Video Activities
- Reading Activities
- Document Activities
- Enrolling People in a Course
- Bulk Enroll With CSV
- Roles in a Course
- How Progress Is Tracked
- Course Reports
- Knowledge Base Overview
- Libraries and Folders
- Uploading Content Items
- Library Visibility
- AI Agent per Library
- My Courses
- Platform People
-
- Project Structure Template
- Table of Contents (TOC)
- Content Folder
- Introduction to Markdown
- Markdown Basics
- Markdown Lists
- Markdown Links
- Markdown Images
- Markdown Code
- Markdown Tables
- Markdown Equations
- Markdown Videos
- Markdown Embedded HTML
- VS Code Snippets
- Introduction to Styles
- Framed Narrations
- Markdown Configuration
- Editor Setup
-
-
-
Publishing These Docs
TL;DR; The whole docs site is republished by clicking one button, which re-imports everything from a configured GitHub branch. One bad page fails the entire import and changes nothing — it never leaves the site half-updated.
Before you start
Nothing — this is a button in the tools menu, under Admin Tools, not a form.
Steps
Select Update Docs. It re-imports every page this site lists from a configured repository and branch.
Every page is fetched and checked before anything is written — a page listed in the nav whose file doesn't exist, an entry without a name, an entry with neither a page nor children, two entries that would share one URL, or nesting deeper than five levels. If even one fails that check, nothing is written at all: the site keeps showing exactly what it showed before you clicked the button.
Only once every page passes does the replacement happen, and it happens as a single all-or-nothing step — there's no partially-updated state where some pages are new and others are stale.
If the import fails for any reason, you'll see "Documentation could not be updated right now. The team has been notified," and a failure report goes out by email with enough detail to find the page that broke it.
What this doesn't do
There's no scriptable way to trigger this — it's a button behind a platform-owner sign-in, not an API endpoint anyone can call directly.
The pre-check is about the nav and the files, not the prose: a page with a typo in its front matter still imports, it just renders oddly. Read the page on the site after publishing if you changed its header.
Common issues
"Documentation could not be updated right now." — Something in the source failed the pre-check: a page listed in the nav that doesn't exist yet, or two entries resolving to the same URL, are the usual causes. Nothing was changed; check the failure email for which page it was.