Tool selection

As explained in The Art of AI Product Development

Tool selection is the process by which an agent chooses an appropriate external tool for a task based on tool descriptions and the current request.

Tool selection is how an agent chooses an external capability for the current request. It depends on clear tool descriptions and remains an important failure point. section 9.1.4

Design tool selection around a small, clearly differentiated set:

• Limit an agent to roughly four or five tools; use multiple specialized agents when broader coverage is necessary. section 9.1.4
• Describe each tool’s purpose, input format, and appropriate situations precisely. section 9.1.4
• Add guidance for cases where a tool should or should not be used, then test for missed and excessive use. section 9.1.4
• Treat write access as a higher-impact capability requiring failure analysis, security attention, and guardrails. section 9.1.4

Matching requests to tools

A biomedical question can point to PubMed, financial news about a public company to Yahoo Finance News using its ticker, and a person-focused video request to YouTube Search with a name and result limit. section 9.1.4

Figure 9.7 The agent’s process for using a tool

Use the process diagram to inspect tool use as a sequence of steps and potential failure points. section 9.1.4

How The Art of AI Product Development covers it

How deep it goes

Explains description-driven selection, missed or excessive use, tool-count limits, and write-access risks, without detailing model internals.

Examples

Uses PubMed, Yahoo Finance News, and YouTube Search to show domain, input, and output distinctions.

What it assumes

Assumes familiarity with prompt engineering and agentic workflows while teaching tool descriptions from concrete examples.

What you can do after

Diagnose/debug with it by refining descriptions when agents miss tools or invoke them unnecessarily.

How it connects

Links tool selection to prompt engineering, specialized agents, security, and guardrails.

liveBookmanning.com

Read The Art of AI Product Development on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Build an AI Agent (From Scratch)Tool selection is the process by which a language model determines whether a tool is needed and chooses the appropriate available tool.

Compare how 2 books explain tool selection →

Related concepts

• Retrieverused by · A retriever is a software component that searches or queries an information source and returns items relevant to a query for use by downstream processes.
• semantic similarityuses · Semantic similarity is the degree to which two items or representations correspond in meaning, regardless of differences in wording or form.
• evaluationused by · Evaluation is the systematic assessment of a system, result, or approach against relevant criteria, evidence, or objectives.
• similarity searchuses · Similarity search is a retrieval technique that finds or ranks items by comparing their representations with a query according to a similarity or distance measure.
• AI agentuses · An AI agent is a computational system that uses artificial intelligence to interpret information, select actions, and pursue goals through interaction with an environment.
• tool usageuses · Tool use is an AI system's invocation of external software functions, services, or other capabilities to obtain information or perform actions.

This explainer was generated by Manning's knowledge graph from the text of The Art of AI Product Development. The section links point to the parts of the book it draws on.