Test and improve the tools your AI assistant relies on

Lets you chat with your AI assistant to test its connected tools and spot ones that need better descriptions.

How the work actually flows

It branches. Exactly one path is taken; runs once per each connected tool.

Pattern: Exclusive Choice (4) · Simple Merge (5) · Multiple Instances without Synchronization (12)

flowchart TD trig(("user sends test question")):::human s0["assistant selects a tool"]:::task s1["draft final answer with ai"]:::svc s2[["review each tool setup"]]:::mi s3["generate improvement report"]:::task trig --> s0 s1 -->|"one per each connected tool"| s2 s2 --> s3 gx{"× which tool to call"}:::gate s0 --> gx p00["call time tool"]:::task gx -->|"time lookup"| p00 p10["call news tool"]:::task gx -->|"news lookup"| p10 jn{"○ answer drafted"}:::gate p00 --> jn p10 --> jn jn --> s1 out[/"working assistant and tool report"/]:::out pay{{"clearer tools and better answers"}}:::pay s3 --> 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 stepAn outside serviceRuns once per itemA personOne path onlyPaths rejoinResultPayoff
Build size
Advanced

A larger build with multiple systems, AI reasoning, and custom rules.

Business functions
AI Agents & Autonomous SystemsAI Chatbots & Assistants
Connects
Google Gemini

The problem it solves

Your AI assistant can only use the tools you've given it, but if those tools are poorly described or missing, it gives vague or wrong answers. Figuring out which tool descriptions need work usually means manually testing every possible question.

Who it fits

A team building an AI assistant that relies on a set of connected tools and wants to make sure those tools actually work well.

How it works

  1. You send the assistant a test question, such as the current time or recent news
  2. The assistant decides which connected tool can answer it and calls that tool
  3. Google Gemini reads the tool's response and writes a final answer
  4. The system reviews each tool's setup against best practices
  5. You get a report of which tools need clearer descriptions
What you get

Tool descriptions flagged as unclear before they cause errors

You get a clear report on which of your AI assistant's tools need better descriptions, so it performs more reliably.

What you get

A working chat assistant plus a review of which of its tools need to be fixed or clarified.

What you need

A Google Gemini API key and hosting for your AI assistant's tool server.

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