Tool Selection: How the LLM Decides When and Which Tool to Call

~15 min read

The LLM picks a tool by reading its name and description against the current task — which means prompt/description design directly determines whether the right tool gets called at the right time.

Tool Selection: How the LLM Decides When and Which Tool to Call is a Pro topic

Sign in, then upgrade to Pro or Power to unlock this topic and the full AI Engineering curriculum.

Key points

  • The model decides whether/which tool to call by matching each tool's name and description against the current request
  • Tool selection quality is almost entirely a function of description quality, not the underlying implementation
  • More tools doesn't mean better results — unnecessary or overlapping tools confuse selection and reduce efficiency
  • Descriptions should specify WHEN to use a tool, not just what it does
  • Keep the active tool list scoped to what's genuinely relevant for the current context, not every tool ever defined