> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vozagent.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Uploading Documents

> How to upload PDFs and other document files to your knowledge base for your AI assistant to reference.

If you have documents like price lists, brochures, policy guides, or manuals, you can upload them directly to your knowledge base. VozAgent extracts the text content and makes it searchable so your assistant can reference it during phone calls.

***

## Supported File Formats

VozAgent supports the following document formats:

| Format         | Extension       | Description                                   |
| -------------- | --------------- | --------------------------------------------- |
| **PDF**        | `.pdf`          | The most common format for business documents |
| **Word**       | `.docx`, `.doc` | Microsoft Word documents                      |
| **Plain text** | `.txt`          | Simple text files                             |
| **HTML**       | `.html`         | Web page files saved locally                  |
| **EPUB**       | `.epub`         | Electronic publication files                  |

***

## File Size Limit

The maximum file size is **10 MB**. If your file exceeds this limit, you'll see an error message: "File size must be less than 10MB."

For large documents, consider breaking them into smaller files or using the [text content](/help/knowledge-base/adding-text-content) option to paste the most important sections manually.

***

## How to Upload a Document

### From the Knowledge page

1. Go to **Knowledge** in your dashboard sidebar

2. Click the **Add Knowledge** button

3. Select **Document** from the resource type options

4. Enter a **Title** for the document (optional — if left blank, the filename will be used)

5. Click the upload area or the **Choose File** button to select your file

6. Once your file is selected, you'll see a green checkmark with the filename and file size

7. Click **Add Knowledge** to upload

The helper text reads: "Upload a document file. Supported formats: PDF, DOCX, TXT, HTML, EPUB."

### From the Upload PDF dialog

If you're using the legacy document view, you may see the **Upload PDF File** option:

1. Click **Add Document** and select **Upload PDF File**

2. Click the upload area or the **Browse files** button to select your PDF

3. Once selected, the file appears with a green checkmark, the filename, and the file size

4. Enter a **Name** for the document (it auto-fills with the filename minus the `.pdf` extension)

5. Click **Upload**

The upload area displays "Drag and drop files here" and "Max. 10MB" to guide you.

***

## What Happens After Upload

After you upload a document:

1. The item appears in your Knowledge list with a **PDF** badge and a file icon
2. The status shows **Scanning** while VozAgent extracts and indexes the text content
3. Once processing completes, the status clears and the content is ready

Processing usually takes a few seconds to a minute depending on the document size and complexity.

***

## Viewing Uploaded Document Content

To see what text was extracted from your document:

1. Find the document in your **Knowledge** list
2. Click the **eye icon** (View content) button

A dialog opens showing the extracted text content. The dialog title reads "Document Training Resource" and displays the date the document was added.

The content is rendered as formatted text. Headings, paragraphs, and lists from the original document are preserved where possible.

***

## Processing Status

| Status              | What it means                                                                                                |
| ------------------- | ------------------------------------------------------------------------------------------------------------ |
| **Scanning**        | The document is being processed and the text is being extracted.                                             |
| *(No status badge)* | Processing is complete. The content is searchable by your assistant.                                         |
| **Error**           | Something went wrong during extraction. The file may be corrupt, password-protected, or contain only images. |

If a document shows an error, you'll see a brief error message below the item explaining what went wrong.

***

## Tips for Uploading Documents

* **Text-based PDFs work best** — PDFs that contain actual text (not scanned images) produce the best results
* **Keep documents focused** — a 5-page FAQ document is more useful than a 200-page operations manual
* **Use descriptive titles** — name your documents clearly (e.g., "2024 Service Pricing" instead of "document1") so you can find them easily later
* **Check the extracted content** — after uploading, use the view button to verify the extracted text looks correct
* **Scanned documents may not work** — if your PDF is a scan of a printed page (image-only), the text extraction may fail or produce poor results

***

## Editing and Deleting Uploaded Documents

Uploaded documents (PDFs and other files) **cannot be edited** after upload. If you need to update the content, you'll need to delete the existing item and upload the updated file.

To delete an uploaded document:

1. Find the item in your **Knowledge** list
2. Click the **trash icon** (Delete) button
3. Confirm the deletion in the dialog

If the document is linked to any assistants, the confirmation dialog will warn you and show how many assistants are affected. You'll need to type **DELETE** to confirm.

For more details on managing items, see [Managing Knowledge Items](/help/knowledge-base/managing-knowledge-items).

***

## Related Articles

* [Knowledge Base Overview](/help/knowledge-base/knowledge-base-overview)
* [Adding Website Content](/help/knowledge-base/adding-website-content)
* [Adding Text Content](/help/knowledge-base/adding-text-content)
* [Managing Knowledge Items](/help/knowledge-base/managing-knowledge-items)
* [Assistant Knowledge Base](/help/assistants/assistant-knowledge-base)