Each document in DocuTrain is its own product. It has a title and a slug, but it also has its own welcome message, its own access rules, its own AI personality, its own UI toggles, its own embed widget, its own quizzes, and its own audit history. The Document Editor — a tabbed modal inside the admin dashboard — is where all of that lives.
This post walks through every tab of the editor and what each setting does.
The shape of the editor
The Document Editor is grouped into four sections in the sidebar:
- Content — what the user sees and reads
- Configuration — access, UI options, attachments, embed, and integrations
- Learning — quizzes, retraining, history, and logs
- Admin — advanced details (super-admin only)
Content section
Overview
The Overview tab is the dashboard for the document. It shows processing status, how much content was indexed, recent activity, and links to retrain or regenerate the abstract. Use it as the "is this document healthy?" view.
Configuration Checklist
A checklist of recommended setup steps — intro message written, cover image set, access level chosen, attachments uploaded, abstract generated. Helpful for owner admins who are setting up many documents in a row.
Images and Videos
Two galleries you can attach to a document:
- Image gallery (Lightbox) — bulk-upload images that users can browse in a lightbox viewer. Useful for medical diagrams, product photos, or step-by-step photography. Turn it on under UI Options, then configure the images here.
- Video gallery — attach video URLs (YouTube, Vimeo, or hosted MP4s) that appear in a video gallery widget inside the chat UI. Same pattern: enable the tool in UI Options, then add links here.
Messages
The Messages tab houses the intro message — the HTML-formatted welcome block that greets users when they open the chat. It supports a rich-text editor with formatting, lists, links, and inline HTML.
If you've generated an abstract for the document (see the AI post in this series), DocuTrain can auto-populate the intro message from the abstract, including a "Sample questions you might try" section that the AI assembles from the document's content.
Disclaimer
A modal that pops over the chat the first time a user opens the document, asking them to acknowledge a disclaimer before chatting. Configurable per document:
- Toggle to show or hide the modal
- Custom disclaimer text up to 10,000 characters (HTML allowed)
- A medication-warning override that inherits from the owner's default, with a per-document override
- A live preview of how the modal will appear
Critical for medical, legal, or regulated use cases where users need to acknowledge that the AI's responses are informational and not a substitute for professional advice.
Helpers (Smart Action Buttons)
A configurable set of one-click suggestion buttons that appear above the chat input. Each helper has a label, an emoji, and a prefilled prompt — when a user clicks it, the prompt is sent to the chat. Great for surfacing common questions like "Summarize the key points" or "What are the contraindications?".
Abstract
The Abstract tab manages the AI-generated executive summary of the document. The system can:
- Generate a fresh abstract from the indexed content
- Regenerate on demand (e.g., after retraining)
- Track when it was last generated and whether the latest run succeeded
- Show a clear status (idle, in progress, or failed) so you know if something needs attention
The abstract feeds into the intro message, the document's search and social previews, and the sitemap.
Configuration section
Settings
The headline tab. Toggles and dropdowns that control how the document behaves on the platform:
| Setting | What it controls |
|---|---|
| Published | Whether the document is live. Unpublished documents don't appear in lists or accept new chats. |
| Access level | Who can chat with the document (covered in the next post). |
| Hide from owner selector | Keep the document out of the owner's document dropdown even when it's published. |
| Discoverable | Allow the document to appear in DocuTrain's public search directory. |
| Include in sitemap | Include in the SEO sitemap (for suitable public documents). |
| Context depth cap | Optional super-admin cap on how much source material each answer can draw from. |
UI Options
This is where you customize the chat page itself. Every toggle is independent:
- Show a dropdown of the owner's other documents
- Show a keyword cloud based on the document content
- Show the downloads sidebar
- Show inline references / citations in answers
- Show recent user questions (anonymized) as suggestions
- Show country flags in analytics views (where available)
- Show the image gallery entry point
- Show the video gallery entry point
- Show the quiz button (where your plan allows)
- Show the abstract button (where your plan allows)
- Let users switch between sidebar and classic layout
- Show a "Back to [Owner]" link at the top
- Use the owner's chat-assistant avatar on this document
Together, these let you turn a single document into anything from a barebones FAQ to a fully-featured training portal with quizzes, downloads, and a video library.
Attachments
Manage downloadable files and external links that appear in the downloads sidebar. Each attachment has a title, a URL or uploaded file, a display order, optional file metadata (size, type), and space for copyright acknowledgement.
Embed & Share
DocuTrain generates a one-line <script> snippet that embeds the document as a floating chat widget on any external website. You can customize:
- Widget button color
- Header color and header text color
- Position (left or right)
- Custom title and custom greeting message
This is how DocuTrain customers drop their AI chatbot onto a marketing site or product page without needing to build any UI themselves.
Telegram
Per-document Telegram bot integration. Connect a bot, and the document becomes accessible via Telegram chat — same answers and citations, same access rules, same conversation history, just a different surface.
Learning section
Quiz
The Quiz tab manages AI-generated quiz questions for the document. The system can generate a configurable number of multiple-choice questions in batches for speed, and you can review and curate them before publishing.
When quizzes are enabled for the document and questions exist, users see a "Quiz" button in the chat UI and can take the quiz interactively.
Retrain
A document's searchable segments are computed when it's processed. If you change the source file or processing options, use Retrain — DocuTrain re-runs the full processing flow from extraction through indexing.
History
Every retrain run is recorded. The History tab is your audit trail: what was processed, when, how much content resulted, and whether it succeeded.
Logs
The Logs tab shows step-by-step processing output. Use it when something looks stuck or failed and you need more detail than the Overview provides.
Admin section
Metadata (super-admin only)
A raw JSON view for advanced troubleshooting — feature flags, processing options, helper configurations, and Telegram bindings. Most teams never need this; it's there when support asks for a deeper look.
Why so many knobs?
Because every document is doing a different job. A public marketing FAQ wants a friendly intro message, no disclaimer, no quiz, and a simple chat UI. A clinical guideline wants a hard disclaimer, medication warnings, references shown, and access restricted to verified members. A training module wants quizzes, an image gallery, and a custom prompt that nudges the AI into a teaching tone.
DocuTrain doesn't try to guess which one you're building — it lets you configure each document to its purpose. The next post covers the first big lever in that configuration: who can see and chat with the document.
→ Next: Access Control: Public, Registered, Passcode, and Beyond