Skip to content

Getting Started with the AgentQ Deep Dive

AgentQ is the AI assistant built into Qualytics. It answers questions about your workspace, explains what it finds, and carries out platform actions on your behalf, all from a chat.

This section is the reference behind that. It explains the machinery rather than the clicks: what AgentQ is allowed to see, how it reaches your data, which providers can run it, what the protocol underneath it is, and where its limits sit. For the click-by-click instructions, see the How-tos.

Where to Start

Read How It Works first. It sets out the surfaces AgentQ appears on, what happens during a single turn, and the two limits that decide what it can reach. Everything else here goes deeper on one piece of that picture.

From there the section falls into four groups.

  • The protocol and the tools. MCP, AgentQ in Action, Tool Catalog, and Semantic Reporting cover what AgentQ can actually do and how an external client reaches the same capabilities.
  • Governance. Access Controls, AgentQ Audit, and Permissions cover who may use AgentQ, how much of your data it may send to the model, and what is recorded.
  • The provider. Supported AI Providers and Amazon Bedrock Authentication cover what can run AgentQ and how to authenticate to it.
  • Using it well. Conversations, Responses & Context, AgentQ Limits, Examples, and Best Practices cover writing good prompts, the ceilings you will meet, and worked scenarios. To build your own applications on the same capabilities, see the AgentQ API.

What This Section Covers

  • How It Works


    Where AgentQ appears, how a turn runs, and what it is allowed to see.

    How It Works

  • The Chat Interface


    Every control in the full-page and floating chat, and what the input accepts.

    The Chat Interface

  • MCP


    What the Model Context Protocol is, how it works, and why it matters.

    MCP

  • AgentQ in Action


    How Qualytics implements MCP, with its endpoint, tools, and tool step labels.

    AgentQ in Action

  • Tool Catalog


    Every tool AgentQ can call, what each one does, and what it shares with the model.

    Tool Catalog

  • Semantic Reporting


    The structured query interface AgentQ uses to count, filter, and inspect platform resources.

    Semantic Reporting

  • Access Controls


    How an administrator chooses what chat may share with the model, and what each level unlocks.

    Access Controls

  • AgentQ Audit


    What each audit entry holds, how cost estimates work, and what the period summary reports.

    AgentQ Audit

  • Supported AI Providers


    Every provider you can connect, what the Beta badge means, and which ones take file attachments.

    Supported AI Providers

  • Amazon Bedrock Authentication


    The three ways to authenticate to Bedrock, and what an IAM role setup expects.

    Amazon Bedrock Authentication

  • Conversations, Responses & Context


    How to write prompts, read AgentQ responses, and work with context-aware chats.

    Conversations, Responses & Context

  • AgentQ Limits


    Rate limits, token usage, timeouts, SQL constraints, and scope constraints.

    AgentQ Limits

  • Examples


    Worked scenarios showing what you ask AgentQ, what it does, and the shape of the answer.

    Examples

  • Best Practices


    Prompt design, cost management, guardrail behavior, rate limits, and async operation patterns.

    Best Practices

  • Permissions


    The user roles behind chatting with AgentQ, configuring it, and reading the audit.

    Permissions