Adding plain text

Last updated: February 12, 2026

Add plain text directly to your chatbot's knowledge base. This is useful for quick additions, custom content that does not exist in a file or on a website, or when you want to type specific knowledge for your chatbot without creating a document first.

Where to find text training

Select your chatbot, then navigate to the Training tab. In the left sidebar, choose Text. This is where you manage all text-based training sources.

Text sidebar in Training tab

How to add text

  1. Click the Add New Training: Text button at the top of the page.

Add New Training Text button

  1. Fill in the form fields:

    • Label -- A name to help you identify this text entry. The label is not used for training and will not affect chatbot responses. It is only for your own reference.
    • Text -- The content you want your chatbot to learn from. Minimum length is 5 characters, maximum length is 100,000 characters.
  2. Click Start Training to begin the training process.

Text form filled with sample data and the Start Training button highlighted

How text processing works

After you click Start Training, ChatLab processes your text in the background:

  1. Processing -- ChatLab splits the text into chunks optimized for retrieval.
  2. Training -- The chunks are converted into embeddings and indexed into your chatbot's knowledge base.

Once training is finished, your text entry appears in the Text sources list with a green "Training completed" status. Each entry shows the character count and the date it was last updated.

Training completed text list showing the Edit button

Editing existing text

To update a text entry, click the Edit button next to it. The edit form opens inline with the current label and text pre-filled. Make your changes and click Update to retrain with the new content.

Edit form for an existing text entry

When to use text training

  • Internal knowledge -- Company policies, procedures, or guidelines that are not published on your website
  • Quick updates -- Time-sensitive information like holiday hours, temporary promotions, or event details
  • Custom responses -- Specific phrasing or answers you want the chatbot to use for particular topics
  • Supplementary content -- Additional context to complement what was already trained from website pages or uploaded files

Related articles