AI Agents

Hitachi iQ Studio User Guide

Version
1.0.x
Audience
anonymous
Part Number
MK-26HIQS001-00
ft:lastEdition
2026-03-31

Agents in iQ Studio are decision-making components that process user queries and orchestrate tasks across tools and knowledge bases. They are configurable and task-oriented assistants. Each agent can be customized with its own purpose, tools, retrieval sources, and behavior. Agents interact with external systems through MCP servers and use knowledge bases as content repositories. MCP servers provide tools such as APIs, connectors, and retrieval functions. Knowledge bases supply curated, authoritative content and use vector-embedded chunks for efficient semantic retrieval.

iQ Studio provides predefined templates for four distinct agent types: Worker, Supervisor, Time‑Travel, and Researcher. Each template is designed for a specific operational role within the flow, enabling users to select an agent structure that aligns with their orchestration requirements.

The following table compares the capabilities of each agent type.
Agent Primary role
Worker Agent Perform specialized and tool-driven tasks using preconfigured tools.
Supervisor Agent Coordinates multiple Worker Agents to manage workflows and compile results.
Deep Researcher Agent Performs in-depth and use-case-driven analysis and generates detailed reports.
Time Travel Agent Can retrieves information from earlier versions of a document.