32dots HEIDELBERG AI
Curriculum › Flowise course › flowise-02
Session 2 medium

Give the agent tools: a Chatflow assistant

LESSONLesson 2 · ~25 min

🎯Goal. Build a single-agent chat assistant that can call tools and use your documents, not just chat.

▶ Try this prompt

When the user asks a factual question, search the web and cite your source; otherwise answer from the uploaded documents.

Flowise has three builder types — Assistant, Chatflow, and Agentflow. A Chatflow is a single agent with tool calling and RAG; that's what you use here.

  1. 1Start a Chatflow — Flowise's single-agent builder that supports tool calling and RAG together. Wire in your chat model as the agent's brain.
  2. 2Add a tool node and connect it to the agent so it can take an action (Flowise ships 100+ integrations across major LLMs, embedding models, and vector databases to choose from).
  3. 3Open the chat panel and ask a question that needs the tool. Watch the assistant decide to call it, then answer — a working tool-using agent built entirely by connecting nodes.
A Flowise Chatflow chat assistant answering questions in an embeddable chat panel with a 'Type your question' input
A Chatflow assistant in Flowise's chat panel — a single agent with tool calling and RAG, testable in the browser before you publish it. Source: https://flowiseai.com

✓You'll see. A chat assistant that picks up a tool when it needs one and otherwise answers from your documents — one agent, real capabilities.

💳Cost. Runs against your 100 predictions/month on the Free plan; tool-calling turns spend extra LLM tokens billed by your provider.

💡Takeaway. A Chatflow is one capable agent: give it tools and documents and it decides when to use each.