The assistant is automatically enabled on Pro and Custom plans.

​
About the assistant

The assistant answers questions about your documentation through natural language queries. It is embedded directly in your documentation site, providing users with immediate access to contextual help. The assistant uses agentic RAG (retrieval-augmented generation) with tool calling powered by Claude Sonnet 4. When users ask questions, the assistant:
  • Searches and retrieves relevant content from your documentation to provide accurate answers.
  • Cites sources and provides navigable links to take users directly to referenced pages.
  • Generates copyable code examples to help users implement solutions from your documentation.
Each message sent to the assistant counts toward your plan’s message allowance. If you exceed your allowance, additional messages incur overage charges. You can set spending limits or disable the assistant if you reach your message allowance. You can view assistant usage through your dashboard to understand user behavior and documentation effectiveness. Export and analyze query data to help identify:
  • Frequently asked questions that might need better coverage.
  • Content gaps where users struggle to find answers.
  • Popular topics that could benefit from additional content.

​
Configuring the assistant

The assistant is enabled by default for Pro and Custom plans. You can manage the assistant from your dashboard, including enabling or disabling it, configuring response handling, and setting a spend limit.

​
Enable or disable the assistant

Toggle the assistant status to enable or disable the assistant for your documentation site.

​
Response handling

Configure how the assistant responds when it cannot find answers in your documentation. You can redirect unanswered questions to your support team and enable the assistant to search external websites.
  • Assistant Deflection: When enabled, the assistant directs users to your support team for questions that it cannot answer.
  • Search Sites: The assistant can search specified external sites for additional context when answering questions.
    Site search is in beta. To enable it for your documentation site, contact our sales team.

​
Spend limit

Set a spend limit to control what happens if you reach your message allowance. By default, the assistant will continue to answer user questions after you reach your message allowance, which incurs overages. In the Billing tab of the Assistant Configurations page, you can set a spend limit for assistant messages beyond your allowance. When you reach your spend limit, the assistant is disabled until your message allowance resets. You can also set usage alerts to receive an email when you reach a certain percentage of your spend limit.

​
Using the assistant

Users can access the assistant in three ways:
  • Keyboard shortcuts: ? or Command + I (Ctrl + I on Windows)
  • Assistant button next to the search bar Search bar and assistant button in light mode.
  • URLs with ?assistant=open appended will open the assistant when the page loads. For example, https://mintlify.com/docs?assistant=open.
Both methods open a chat panel on the right side of your docs. Users can ask any question and the assistant will search your documentation for an answer. If no relevant information is found, the assistant will respond that it cannot answer the question.

​
Making content AI ingestible

Structure your documentation to help the assistant provide accurate, relevant answers. Clear organization and comprehensive context benefit both human readers and AI understanding.

Structure and organization

  • Use semantic markup.
  • Write descriptive headings for sections.
  • Create a logical information hierarchy.
  • Use consistent formatting across your docs.
  • Include comprehensive metadata in page frontmatter.
  • Break up long blocks of text into shorter paragraphs.

Context

  • Define specific terms and acronyms when first introduced.
  • Provide sufficient conceptual content about features and procedures.
  • Include examples and use cases.
  • Cross-reference related topics.
  • Add hidden pages with additional context that users don’t need, but the assistant can reference.

​
Exporting and analyzing queries

Review and export queries from your dashboard to understand how people interact with your documentation and identify improvement opportunities. Some ways that analyzing queries can help you improve your documentation:
  • Identify content gaps where frequent queries receive insufficient answers.
  • Discover user behavior patterns and common information needs from themes and patterns in queries.
  • Prioritize high-traffic pages for accuracy and quality improvements.
You can explore queries from your dashboard, but to get more powerful insights we recommend exporting a CSV file of your queries, responses, and sources to analyze with your preferred AI tool.
  1. Navigate to the assistant page in your dashboard.
  2. Select Export to CSV.
  3. Analyze the exported data using your preferred tool.

Sample analysis prompts

  • Summarize the most common themes of the queries.
  • List any queries that had no sources cited.
  • Find patterns in unsuccessful interactions.