An LLM invoking external functions to act beyond text generation.
1 min readupdated 2026-07-04
/ quick answer
Tool use is the pattern where a model returns a structured call (name + arguments) that the runtime executes, then feeds the result back into the conversation for the next step. An LLM invoking external functions to act beyond text generation.
An LLM invoking external functions to act beyond text generation. Tool use is the pattern where a model returns a structured call (name + arguments) that the runtime executes, then feeds the result back into the conversation for the next step. In practice: A model returns { tool: 'search_web', args: { q: 'Notion pricing 2026' } }; the runtime runs the search and returns hits to the model. This dictionary node is part of the Onexial knowledge graph and links to related concepts, workflows and tools below.
Definition
Tool use is the pattern where a model returns a structured call (name + arguments) that the runtime executes, then feeds the result back into the conversation for the next step.
Example
A model returns { tool: 'search_web', args: { q: 'Notion pricing 2026' } }; the runtime runs the search and returns hits to the model.
Tool use is the pattern where a model returns a structured call (name + arguments) that the runtime executes, then feeds the result back into the conversation for the next step.
What is an example of Tool Use?
A model returns { tool: 'search_web', args: { q: 'Notion pricing 2026' } }; the runtime runs the search and returns hits to the model.
Why does Tool Use matter for AI and automation?
An LLM invoking external functions to act beyond text generation. It connects to the workflows, prompts and tool stacks linked on this page, so you can move from definition to execution without leaving Onexial.