AI Agent Management
Data Sources
Add and maintain the business knowledge your chatbot uses to answer customers.
Documentation Index
Fetch the complete documentation index at: /llms.txt
Use this file to discover all available QwryAI pages before exploring further.
Supported Sources
Data Sources is where you provide the content the chatbot should trust. Keep this content current and specific to the workspace or storefront you are supporting.
- Website pages
- Help center articles
- Product pages
- Shipping and return policies
- FAQs
- Documents and internal support notes
| Source type | Best use |
|---|---|
| Website pages | Public support pages, landing pages, policies, and product guides. |
| Documents | Internal support playbooks, PDFs, and structured knowledge files. |
| Text or Q&A | Short exact answers for repeated questions or policy exceptions. |
| Product content | Catalog and store information customers ask about before buying. |
- Keep sources tied to the selected workspace.
- Remove outdated content that could produce wrong answers.
Add a Data Source
- Open your workspace.
- Select the chatbot you want to train.
- Open Data Sources.
- Choose the source type.
- Add the URL, file, text, or document content.
- Start processing.
- Wait until processing is complete before testing.
Preview and Test
After a source is processed, ask questions in Playground that the new content should answer. This confirms that the source is useful and written clearly enough for retrieval.
- Ask the exact question the new source should answer.
- If the answer is weak, improve the source text before changing model settings.
Maintenance
- Review sources after policy, pricing, product, or shipping changes.
- Delete duplicate or obsolete content.
- Use clear headings and direct language.
- Add missing answers when Knowledge Gaps show repeated questions.