What Is Tool Use in Prompts and How LLMs Parse Function Calling Schemas
Tool use in LLMs is function calling — the model emits structured JSON naming a tool and arguments. Your code executes; the model reasons over the result.
Tool Use in Prompts
Tool use in prompts lets LLMs call external functions, APIs, and tools by embedding schema definitions directly in the prompt context.
The model reads available tool signatures, decides when to invoke them, and returns structured calls your application executes — unlocking real-world actions like querying databases, running code, and fetching live data that go beyond text generation. Also known as: Function Calling, Tool Calling
What this topic covers
- Foundations — Tool use in prompts transforms LLMs from text generators into action-capable systems by embedding function schemas directly in the context.
- Implementation — The practical guides cover writing effective tool descriptions, structuring function calling schemas, and handling the failure modes that appear when models select the wrong tool or return malformed parameters.
- What's changing — Function calling capabilities are evolving rapidly across model providers, with benchmark performance diverging sharply from real-world reliability.
- Risks & limits — When LLMs call external APIs, the consequences are real and often irreversible.
This topic is curated by our AI council — see how it works.
1
Understand the Fundamentals
Tool use in prompts transforms LLMs from text generators into action-capable systems by embedding function schemas directly in the context. Understanding how models parse and invoke those schemas is what separates reliable integrations from brittle ones.
Concepts covered
Context Costs, Wrong Parameters, and the Hard Limits of LLM Tool Calling
LLM tool schemas vary by provider and add hidden token overhead. Task accuracy peaks near 77% (BFCL v3, 2026) — well below 99%+ schema conformance.