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

# Source Management — Audit, Filter & Bulk-Edit Training Data

> Search, filter, and bulk manage your InsiteChat AI chatbot's training sources. Re-sync, delete, or audit website URLs, files, Q&A pairs, and cloud integrations.

The **InsiteChat Sources tab** provides tools for managing all your AI chatbot's training data in one place — website URLs, uploaded files, text snippets, custom Q\&A pairs, and cloud integrations like [Google Drive](/integrations/google-drive), [Notion](/integrations/notion), and [Dropbox](/integrations/dropbox). Filter by status, search by name or URL, and run bulk actions (re-sync, delete) without opening each source individually.

## Search & filter

### Search bar

Type in the search bar to filter sources by title or URL. Results update instantly as you type.

### Status filter

Click status buttons to show only sources with a specific status:

| Status         | Meaning                            |
| -------------- | ---------------------------------- |
| **All**        | Show everything                    |
| **Pending**    | Queued for processing              |
| **Crawling**   | Actively fetching pages            |
| **Processing** | Embedding and indexing content     |
| **Done**       | Successfully trained               |
| **Failed**     | Error occurred (hover for details) |

Each status button shows a count of how many sources have that status.

## Bulk actions

### Select sources

* Click the **checkbox** next to any source to select it
* Click the **select all** checkbox at the top to select all visible (filtered) sources
* Click **Clear selection** to deselect all

### Available bulk actions

When one or more sources are selected, a bulk action bar appears:

| Action      | What it does                                                                           |
| ----------- | -------------------------------------------------------------------------------------- |
| **Re-sync** | Queues all selected sources for re-processing. Uses force mode to bypass cache.        |
| **Delete**  | Soft-deletes all selected sources and archives their documents. Requires confirmation. |

<Tip>
  Use bulk re-sync after a major website update to refresh all URL sources at once, rather than clicking re-sync on each one individually.
</Tip>

## Per-source actions

Each source card has two action buttons:

* **Re-sync** (refresh icon) — Re-process this single source
* **Delete** (trash icon) — Remove this source and its trained data

## Error handling

When a source fails to process, an error message is displayed on the source card. Hover over truncated error messages to see the full text.

Common failure reasons:

* Website returned an error (403, 404, 500)
* Content too short (less than 50 characters)
* Monthly page quota exceeded
* File format not supported
* Network timeout during crawl

## Source types

| Type         | Icon      | How it's added                                   |
| ------------ | --------- | ------------------------------------------------ |
| Website URL  | Globe     | Enter URL, optionally with advanced crawl config |
| Text         | FileText  | Paste text content with a title                  |
| File         | File      | Upload PDF, DOCX, PPTX, TXT, MD, CSV             |
| YouTube      | Play      | Enter video URL to extract transcript            |
| Google Drive | HardDrive | Connect account and select files                 |
| Notion       | FileText  | Connect workspace and select pages               |
| Dropbox      | Cloud     | Connect account and select files                 |