Tool Use
Giving a model a set of callable functions so it can act on the world rather than only describe it.
Tool use — also called function calling — is the mechanism by which a Large Language Model invokes external code. The host declares a set of tools, each with a name, a description, and an input schema; the model may emit a structured call instead of prose; the host executes it and returns the result as a new turn. Repeating this is the Agentic Loop.
The tool definition is the model's entire understanding of the capability. Its description is not documentation for humans who happen to be reading — it is the prompt that decides whether the tool gets used correctly, and it is where routing and disambiguation guidance belongs. Because it ships with the implementation it cannot drift, which is why a separate written catalogue of tools is a liability; see Documentation Rot.
Several practical constraints repeat. Input schemas are typically validated by the framework that presents the tools, so code paths that invoke a tool's implementation directly bypass that validation entirely — a security argument resting on "the schema constrains this value" holds only for the routed path. Descriptions of every field are worth requiring mechanically, since a model that must guess a shape will guess. And the set of tools offered is itself context: more tools means more Tokens and more opportunity to pick wrong.
See also6
Hand-picked in the note itself — the neighbours worth reading next.
Structured Output
Constraining a model to emit data conforming to a declared schema rather than free text.
Agents & Language Models11 connections
JSON Schema
A vocabulary for describing the shape of JSON data, and the lingua franca of model tool interfaces.
Agents & Language Models9 connections
Schema Validation
Checking data against a declared shape at runtime, and using the result as the typed value.
Agents & Language Models11 connections
Model Context Protocol
An open protocol for exposing tools, data, and prompts to AI applications through a uniform interface.
Agents & Language Models10 connections
Guardrail
A rule constraining what a model may do, enforced by prompt, schema, or surrounding code.
Agents & Language Models13 connections
Fingerprint
A hash over everything that determined a result, used to detect when the result has gone stale.
Agents & Language Models11 connections
Linked from23
Notes elsewhere in the wiki that reach for this one.
- Agent SkillAgents & Language Models
A packaged, invocable set of instructions that loads into an agent's context on demand.
- Agentic LoopAgents & Language Models
The cycle of model turn, tool call, observation, repeat — the basic mechanism of an AI agent.
- Anchoring EffectMethod
A stated number pulls subsequent estimates toward it, regardless of relevance.
- Constrained DecodingAgents & Language Models
Restricting a model's next-token choices to those a formal grammar permits.
- Context WindowAgents & Language Models
The bounded span of tokens a model can attend to in a single request.
- Evaluation HarnessAgents & Language Models
A repeatable test suite for model behaviour, since prompts and tool descriptions cannot be type-checked.
- FingerprintAgents & Language Models
A hash over everything that determined a result, used to detect when the result has gone stale.
- GuardrailAgents & Language Models
A rule constraining what a model may do, enforced by prompt, schema, or surrounding code.
- HallucinationAgents & Language Models
Fluent, confident output that is not grounded in anything real.
- Human in the LoopAgents & Language Models
Requiring a person's judgement at chosen points in an otherwise automated process.
- JSON SchemaAgents & Language Models
A vocabulary for describing the shape of JSON data, and the lingua franca of model tool interfaces.
- Large Language ModelAgents & Language Models
A neural network trained to predict text, used as a general-purpose instruction-following system.
- Model Context ProtocolAgents & Language Models
An open protocol for exposing tools, data, and prompts to AI applications through a uniform interface.
- NondeterminismAgents & Language Models
The property that identical inputs may produce different outputs, and what it costs to test around.
- PromptAgents & Language Models
The text supplied to a language model to elicit a response.
- Prompt EngineeringAgents & Language Models
Shaping a model's input to make the desired behaviour reliable rather than occasional.
- Prompt InjectionAgents & Language Models
Untrusted content that reaches a model's context and is treated as instruction.
- Schema DriftData & Content
Divergence between a schema and the code, types, or documentation that describe it.
- Schema ValidationAgents & Language Models
Checking data against a declared shape at runtime, and using the result as the typed value.
- Streaming ResponseNetworks & Distribution
Delivering a response incrementally as it is produced rather than all at once when complete.
- Structured OutputAgents & Language Models
Constraining a model to emit data conforming to a declared schema rather than free text.
- System PromptAgents & Language Models
The standing instructions that frame every turn of a conversation with a model.
- TokenAgents & Language Models
The sub-word unit a language model actually reads and writes; the unit of cost and of context.