46.1 Purpose of Pasted-Text Sources

The Sources page includes Paste content directly for situations where the information should be indexed but a normal website crawl is unreliable or inappropriate.

The interface specifically identifies examples such as anti-bot challenges, login walls, or JavaScript-heavy pages that do not crawl successfully.

46.2 How Pasted Text Works

Instead of fetching a remote page, AIUNIFY Chat indexes the text you enter directly.

This provides a practical fallback when you already have the authoritative text but the crawler cannot obtain it.

46.3 Available Fields

The pasted-text form provides:

  • Title (optional)
  • Source URL (optional)
  • Content

46.4 Entering a Title

Add a meaningful Title when possible so the Source and resulting Document are easy to recognize later.

If you leave the Title blank, the system can derive a title from the beginning of the supplied content.

46.5 Optional Source URL

If the pasted material came from a public webpage, enter that original URL in Source URL (optional).

This preserves useful reference context even though the crawler itself is being bypassed.

46.6 Minimum Content Requirement

The Content field displays its current character count and requires at least 50 characters before the Add action becomes available.

The deployed server-side validation also applies a maximum content size for a single pasted-text submission.

46.7 Adding the Source

  1. Open the Agent's Sources page.
  2. Locate Paste content directly.
  3. Enter an optional Title.
  4. Enter an optional original Source URL.
  5. Paste the authoritative text into Content.
  6. Confirm the text is at least 50 characters.
  7. Select Add as source.

46.8 Processing

The content is added as a Pasted Text Source and queued for indexing without going through the normal webpage crawler.

After processing, review the resulting Document and Chunks in the Knowledge workspace.

46.9 Updating Pasted Content

If the underlying information changes, verify that the Agent's indexed copy still matches the authoritative content. Depending on the change, reindex or replace the Source with current material.

46.10 Best Practices

  • Paste clean, useful content rather than entire page layouts.
  • Remove irrelevant navigation or repeated boilerplate where possible.
  • Use a descriptive Title.
  • Include the original URL when it helps trace the information.
  • Verify the resulting Chunks after indexing.
Write Your Comment