32dots HEIDELBERG AI
Session 3 medium

Many agents, one job: build an Agentflow

LESSONLesson 3 · ~30 min

🎯Goal. Orchestrate several agents that hand work to each other to complete a multi-step task.

▶ Try this prompt

From a Start node, route to a Search Agent that gathers facts and a Get Request node that fetches data, then pass both to a Summarize node that writes the final answer.

Use the third builder type, Agentflow — Flowise's canvas for multi-agent orchestration, where each node is a step or an agent and the connections set the order.

  1. 1Create an Agentflow and drop a Start node — the entry point for the run.
  2. 2Add specialised nodes (e.g. a Search Agent, a Get Request tool, a Summarize step) and connect them so work flows from one to the next; you literally see the path the task takes.
  3. 3Run it from the chat panel and watch each agent do its part and hand off. This is the leap from one assistant (Lesson 2) to a coordinated team of them.
A Flowise Agentflow canvas with a Start node branching to a Search Agent and a Get Request node, then connecting to a Summarize node
An Agentflow canvas: a Start node branches to a Search Agent and a Get Request, then converges on a Summarize step — multi-agent orchestration built by connecting nodes. Source: https://flowiseai.com

You'll see. A multi-node workflow where Start branches to a Search Agent and a Get Request, then converges on a Summarize step that produces the answer.

💳Cost. Every agent step is its own model call, so a multi-agent run spends more predictions and tokens than a single Chatflow — plan the flow before you run it repeatedly.

💡Takeaway. Agentflow is for orchestration: when one agent isn't enough, wire several together and watch the hand-offs on the canvas.