Tool Use
The ability of an AI agent to invoke external tools—browsers, APIs, calculators, code interpreters, databases—to accomplish tasks beyond text generation. Tool use is what distinguishes agents from chatbots: a sales agent uses CRM tools, a coding agent uses a terminal, and a finance agent queries accounting software. Modern tool use is implemented through structured Function Calling and increasingly through Model Context Protocol (MCP) for cross-platform compatibility.
Written by Max Zeshut
Founder at Agentmelt · Last updated Sep 9, 2026
Example
An ops agent asked to investigate a payment failure uses tools sequentially: query the database for the failed transaction, call the Stripe API to fetch the underlying error code, search the runbook knowledge base for that error code, and draft a Slack message summarizing the cause and recommended fix. Five tools, four seconds, one structured response—a chatbot without tool use could only restate the question.
Frequently asked questions
- How many tools should an agent have?
- Fewer than you'd think. Production agents work best with 5-20 well-described tools; above ~30 tools, models start mis-selecting (calling the wrong tool, ignoring the right one) at noticeably higher rates. If you have more tools than that, structure them: load tools conditionally via Agent Skills, or split into subagents that each have their own focused tool set.