Experiment with an AI agent that can answer questions and use tools

A simple AI agent setup that answers questions directly or looks things up using connected tools like Wikipedia.

How the work actually flows

It branches. Every path runs; all paths must finish before it continues.

Pattern: Parallel Split (2) ยท Synchronisation (3)

flowchart TD trig(("workflow is run manually")):::human s0["start workflow"]:::task s1["return both results"]:::task trig --> s0 gx{"+ which approach answers question"}:::gate s0 --> gx p00["send question directly to openai"]:::task gx -->|"direct openai path"| p00 p10["agent decides to answer or search wikipedia"]:::task gx -->|"agent tool path"| p10 jn{"+ combine results"}:::gate p00 --> jn p10 --> jn jn --> s1 out[/"two example ai responses returned"/]:::out pay{{"clear example to build own agent"}}:::pay s1 --> out out --> pay classDef task fill:#e7f6fe,stroke:#34b8f0,color:#2c2a29 classDef svc fill:#f6f8fa,stroke:#7c8795,color:#2c2a29 classDef mi fill:#e7f6fe,stroke:#0079a8,color:#2c2a29,stroke-width:2px classDef human fill:#fff,stroke:#0079a8,color:#0079a8 classDef store fill:#f6f8fa,stroke:#0079a8,color:#2c2a29 classDef trig fill:#00a4eb,stroke:#0079a8,color:#fff,font-weight:bold classDef trigtime fill:#00a4eb,stroke:#0079a8,color:#fff,font-weight:bold classDef trigdata fill:#8ad4f5,stroke:#0079a8,color:#06314c,font-weight:bold classDef gate fill:#fff,stroke:#e8a23d,color:#6b4708,font-weight:bold classDef out fill:#1f9d6b,stroke:#167a53,color:#fff,font-weight:bold classDef pay fill:#06314c,stroke:#021f33,color:#fff
A stepA personEvery pathWaits for allResultPayoff
Build size
Standard

A mid-size build with several tools working together.

Business functions
AI Agents & Autonomous SystemsEducation & Training
Connects
OpenAIWikipedia

The problem it solves

You want to see how an AI agent decides when to answer from what it already knows versus when to look something up, but most examples are either too simple or too complex to learn from. You need a clear, working example to build from.

Who it fits

A developer or technical team exploring how to build their own AI agent.

How it works

  1. You run the workflow to start it
  2. One path sends a preset question straight to OpenAI for a direct answer
  3. The other path sends a factual question to an AI agent
  4. The agent decides whether to answer directly or look up the answer on Wikipedia
  5. Both results are returned so you can compare
What you get

Answers you get from comparing two approaches

Ask a question and compare a direct AI answer against one built by an agent that looks things up on Wikipedia.

What you get

Two example AI responses, one direct and one tool-assisted, side by side.

What you need

An OpenAI API key.

We can build this. But should you?

The hard question is not how to build it. It is whether this is the right thing to build first.

That is what a Fractional Chief AI Officer figures out with you, before anyone writes a line of code.

Let's Talk Strategy

Related automations

Back to the AI Playbook