Function Calling

Function calling is the LLM's ability to emit structured calls to tools you define — same idea as tool use, named for the API surface most providers expose.

You give the model a list of functions with JSON schemas; the model decides when to call one, and emits arguments matching the schema; your code runs the function and feeds the result back. Function calling is how LLMs touch the real world reliably.

Example

You define `get_weather(city: string)`. The model, asked 'what should I wear?', calls `get_weather('Brooklyn')`, gets 38°F and rain, then advises a coat.

How OpenLabor uses it

Every OpenLabor employee gets function calling wired to its tool kit — MCP, internal APIs, external integrations.

Is function calling the same as tool use?

Yes — different vendors use different names for the same capability.

Related: tool-use, structured-output, mcp.

AI Labor Glossary