---
title: "Comment by Adam Lewis on \"Tool Design for Agents: Writing Functions a Model Can Actually Use Well\" | Hashnode"
description: "A useful boundary between fixing the description and changing the tool is hard to place from the outside. Your web search example is a description fix. The eigh"
url: "https://hashnode.com/posts/tool-design-for-agents-writing-functions-a-model-can-actually-use-well/6aedc234f62911a4e92fb73e/comment/6af390b09b6de8e126fd4352"
---

[Discussion](https://hashnode.com/posts/tool-design-for-agents-writing-functions-a-model-can-actually-use-well/6aedc234f62911a4e92fb73e)

Comment [Tool Design for Agents: Writing Functions a Model Can Actually Use Well](https://hashnode.com/posts/tool-design-for-agents-writing-functions-a-model-can-actually-use-well/6aedc234f62911a4e92fb73e)

[Adam Lewis](https://hashnode.com/@nark3d)

[2d ago](https://hashnode.com/posts/tool-design-for-agents-writing-functions-a-model-can-actually-use-well/6aedc234f62911a4e92fb73e/comment/6af390b09b6de8e126fd4352)

A useful boundary between fixing the description and changing the tool is hard to place from the outside. Your web search example is a description fix. The eight repeated calls you'd attribute to pagination defaults need the schema changed. When you're watching transcripts and the model keeps making the same mistake, how do you decide the description is underspecified? Both fixes look identical from the outside until you've shipped one and watched it not help.