Managing Documents

Learn how to search, view, download, refresh, and delete your knowledge documents.

Document List View

The Knowledge Base page displays all your organization’s documents in a table format.


Table Columns

Title
Document name, either extracted from the source or user-defined.

Source
Where the content came from (URL, File Upload, Inline Text, or Crawl).

Size
Document size in KB or MB.

Status
Current processing status with color-coded badges.

Agents
Number of agents currently using this document.

Created
When the document was created (e.g., “2 minutes ago”).

Actions
Available operations for the document (Preview, Download, Refresh, Delete).

Search and Filter

Use the search bar at the top of the page to filter documents.

  • Document title

Document Actions

Each document has action icons in the Actions column.

Preview Content

View document content directly in your browser.


Icon: Eye icon

How to use:

  1. Click the eye icon next to the document
  2. A modal opens showing the content
  3. Scroll through to review
  4. Click outside or the X to close

Availability: Text-based documents only (TXT, MD, HTML, inline text)

Use cases:

  • Verify content before attaching to agents
  • Review what information agents will see
  • Check content accuracy

Download Document

Download the original document file to your computer.

Icon: Download arrow icon

How to use:

  1. Click the download icon
  2. File downloads automatically
  3. Open with appropriate application

Availability: All document types

Downloaded file:

  • Same format as uploaded or scraped
  • Original filename preserved
  • Full content included

Refresh Document

Re-scrape URL-based documents to update content with the latest version.

Icon: Circular arrow icon

How to use:

  1. Click the refresh icon
  2. Document status returns to “Pending”
  3. Processing begins automatically
  4. Wait for “Uploaded” status

Availability: URL and Crawl documents only

What happens:

  1. System re-scrapes content from the source URL
  2. New version is created on AI platform
  3. All attached agents are automatically updated
  4. Old version is removed
  5. No manual re-attachment needed

Processing time: Same as initial creation

Use cases:

  • Documentation has been updated
  • Want to capture latest information
  • Fix issues with initial scrape

Retry Failed Upload

Retry processing for documents that failed.

Icon: Retry icon (circular arrow)

How to use:

  1. Find documents with “Error” status
  2. Click the retry icon
  3. Processing starts again
  4. Check error message if it fails again

Availability: Documents with “Error” status only

Common reasons to retry:

  • Temporary network issues
  • Service rate limits
  • Transient API errors

Delete Document

Permanently remove a document from your Knowledge Base.

Icon: Trash icon

How to use:

  1. Click the trash icon
  2. Confirmation dialog appears
  3. Click confirm to delete

Warning: If the document is attached to any agents, you must detach it first before deletion is allowed.

What gets deleted:

  • Document metadata from database
  • Content from cloud storage
  • Document from AI platform
  • All attachments to agents

This action cannot be undone.

Document Details

Click on a document row to view additional details:

  • Full source URL or filename
  • Processing timestamps (created, processed, uploaded)
  • Content hash for deduplication
  • Error messages (if status is “Error”)
  • Crawl settings (for website crawls)

Status Management

Monitoring Active Processing

Documents currently processing show animated status badges. The page refreshes automatically every 5 seconds while any documents are processing.

Handling Errors

If a document shows “Error” status:

  1. Click on the document to view the error message
  2. Check the specific error reason
  3. Click retry if it was a temporary issue
  4. Delete and recreate if the source has problems

Common Error Messages

“Failed to scrape URL”

  • URL is not publicly accessible
  • Website blocks automated scrapers
  • Network timeout occurred

“File too large”

  • File exceeds 10 MB limit
  • Solution: Compress or split the file

“Processing timeout”

  • Content took too long to process
  • Solution: For crawls, reduce max_pages

“Invalid file format”

  • File type not supported
  • Solution: Convert to supported format

Bulk Operations

Multiple File Upload

You can upload multiple files at once:

  1. Click the File tab
  2. Select multiple files using Shift or Cmd/Ctrl + click
  3. All files upload and process independently

Future Features

Coming soon:

  • Bulk delete
  • Bulk attachment
  • Export document list

Organization Isolation

Documents are organization-specific:

  • You only see documents from your organization
  • Cannot share documents between organizations
  • Users see documents based on organization membership

Best Practices

Document Organization

  • Use clear, descriptive titles
  • Keep documents focused on single topics
  • Create separate documents for different subjects
  • Regular cleanup of outdated documents

Content Updates

  • Refresh URL documents regularly (weekly or monthly)
  • Set reminders for important document updates
  • Delete outdated documents rather than leaving them inactive

Agent Management

  • Review which agents are using each document
  • Detach from agents before deleting
  • Test agent responses after attaching new documents

Need Help?

This site uses Just the Docs, a documentation theme for Jekyll.