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

# Upload files

> Upload documents such as PDF, DOC, or CSV files from your computer so your chatbot can learn from them.

Use this guide to add documents from your computer to your chatbot's content. SiteGPT reads the text in each file.

To add files from Notion, Google Drive, Dropbox, OneDrive, or Box, see [Connect apps](/docs/guides/content/connect-apps). For GitHub, see [Connect GitHub](/docs/guides/content/connect-github).

## Before you start

* An active plan. Each plan has a page limit. See [Plans and limits](/docs/reference/plans-and-limits).
* The Super Admin, Admin, or Manager role on the chatbot. Agents cannot add content.
* Files that contain real text. See [What SiteGPT can read](#what-sitegpt-can-read).

Anyone who can use your chatbot can get answers from the files you upload. Only upload content that is safe for them to see.

## Upload files

<Steps>
  <Step title="Open the upload form">
    In the chatbot sidebar, go to **Files & Data Sources** > **Upload Local Files**.
  </Step>

  <Step title="Choose the files">
    Select **Browse File** and choose one or more files. Drag and drop is not supported.
  </Step>

  <Step title="Upload">
    Select **Upload Files**. Wait for the message "All files uploaded successfully! Redirecting..."
  </Step>
</Steps>

You cannot select more files than the pages left in your quota. If you do, the form shows "You can only add N more file(s) with your current plan."

## What SiteGPT can read

* **Formats.** The upload form names PDF, CSV, and DOC as examples. SiteGPT reads the text in the file.
* **Scanned documents.** SiteGPT does not run text recognition (OCR). A PDF that is only scanned images gives no text. Use a PDF with selectable text instead.
* **Large files.** Keep each file under 10 MB. Larger files can fail. Split them into smaller files.
* **Pages of quota.** Each file uses one page for every 2,500 characters of text, and at least one page.

## Check that it works

1. Go to **Files & Data Sources** > **Files List**.
2. Wait for each file to show **Success**.
3. Ask the chatbot a question that one of the files answers.

## Fix text that was read badly

If SiteGPT read part of a file wrongly, you can edit the text it stored.

1. In **Files List**, find the file.
2. Select the pencil icon, **Update Content**, on the file's row.
3. Correct the text in **Document Content**.
4. Select **Update Content**. The message "Document content updated successfully" appears, and SiteGPT trains on your edited text.

<Warning>
  If you resync the file later, SiteGPT reads the original file again. Your edits are then lost. To keep a correction, fix the original file and upload it again.
</Warning>

## Replace or remove a file

SiteGPT does not see changes you make to the file on your computer. To use a new version:

1. In **Files List**, select the old file.
2. Select **Delete N Selected** and confirm.
3. Upload the new version.

Deleting a file removes it from the chatbot's content and frees its pages.

## Troubleshooting

<AccordionGroup>
  <Accordion title="A file shows Failed">
    Point to the **Failed** badge to see the reason. "This file or page format could not be processed" means SiteGPT could not read the file. Save it as PDF or plain text and upload it again.
  </Accordion>

  <Accordion title="A file shows Limit Reached">
    Your plan's page limit was reached. Delete content you do not need, or upgrade. See [Plans and limits](/docs/reference/plans-and-limits).
  </Accordion>

  <Accordion title="Uploads are limited in my HIPAA workspace">
    HIPAA workspaces have their own upload rules. See [HIPAA](/docs/account/hipaa).
  </Accordion>
</AccordionGroup>

Using the API or CLI to upload files? See [Ways to use SiteGPT](/docs/concepts/ways-to-use-sitegpt) for the developer tools.