Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.deflect.in/llms.txt

Use this file to discover all available pages before exploring further.

The Knowledge Base allows your agent to provide accurate, company-specific answers by searching through your documents (PDFs, Markdown, URLs).

​
Adding Knowledge Sources

​
Document Upload

You can upload PDF or Markdown files directly to the agent. Our system will:
  1. Extract the text content.
  2. Split it into manageable “chunks”.
  3. Store them in a vector database for fast retrieval.

​
URL Crawling

Provide a link to your public documentation or website. The system will crawl the pages and index the content.

​
Source Attribution

When the agent answers a question using your knowledge base, it can cite its sources. This builds trust with your users.

​
Adding Metadata to Markdown

To ensure accurate citations for Markdown files, you can add a metadata block at the end of your file: 
## metadata:
```json
{
  "sourceURL": "https://docs.yoursite.com/article-1",
  "title": "How to Configure Settings"
}


## How Retrieval Works (RAG)
We use a technique called **Retrieval-Augmented Generation (RAG)**:
1.  **User Asks**: "How do I reset my password?"
2.  **Search**: The system searches your Knowledge Base for the most relevant text snippets.
3.  **Augment**: These snippets are added to the prompt sent to the AI.
4.  **Generate**: The AI generates an answer based *only* on those snippets.