Tool calling (also called function calling) is the mechanism by which a language model invokes external functions, such as a search, a database query, or an API, by emitting a structured request that an application executes and feeds back as a result. It lets a model act beyond text generation by delegating to deterministic code and live data.
Why it matters
Tool calling is what turns a chat model into an agent: instead of guessing an answer, the model can look something up, run a calculation, or fetch current information and then reason over the real result. It is the runtime foundation of retrieval and of standards like MCP, where servers expose tools the model can choose to call. For knowledge work, it means an assistant can pull from authoritative systems rather than relying on what it happens to remember.
How Pith relates
Pith's MCP server presents its search and retrieval as callable tools, so an AI assistant can decide to query your reading memory and receive cited results in return. This lets the model ground its answers in your saved sources at the moment it needs them, instead of working from a stale copy. The tools return provenance with each result, keeping the assistant's actions traceable.
See also
Last reviewed: 7 June 2026 · Licensed CC BY 4.0 · cite freely with attribution to Pith.