Tool description
A tool description is information specifying a callable tool’s purpose and how an agent or system can invoke it, including its inputs or parameters.
Build an AI Agent (From Scratch)
A tool definition is a structured declaration of a callable tool's name, description, parameters, types, and required arguments for access by a language model.
- How it is used
- Enables an LLM to identify available tools and form structured calls without executing them itself.
- How deep it goes
- Works through schema fields, external execution, Python metadata conversion, and MCP format adaptation.
- Examples
- Uses a calculator schema, generated web-search definitions, and MCP-discovered search tools.
- What you can do after
- Build tool definitions from scratch or generate them from Python functions, then integrate them into a tool-calling flow.
Read how Build an AI Agent (From Scratch) explains tool description →
Vibe Engineering
A tool definition is a machine-readable description of an operation an AI agent can invoke, including its interface and capabilities.
- How it is used
- Frames tool descriptions as agent interfaces that also consume context and require governance.
- How deep it goes
- Defines MCP tool schemas and works through context costs and mitigations without showing a schema implementation.
- Examples
- Uses a five-server coding setup and its measured context breakdown to illustrate schema overhead.
- What you can do after
- Evaluate and understand the context and governance tradeoffs of exposing MCP tools to an agent.
AI Agents in Action, Second Edition
Tool metadata is descriptive information about a callable tool, including its name, purpose, and parameter schema.
- How it is used
- Guides agents in selecting callable tools and supplying valid parameters without duplicating tool details in general instructions.
- How deep it goes
- Explains metadata fields, disambiguation, and parameter guidance through code examples, but does not descend into model internals.
- Examples
- Uses search tools, a keyword-based retrieval tool, SequentialThinking, and a multi-agent research, planning, and filesystem flow.
- What you can do after
- Apply it via an SDK by writing distinct descriptions, documenting parameters, and moving general tool instructions out of the prompt.
Read how AI Agents in Action, Second Edition explains tool description →
Context Engineering
A tool definition specifies a tool's name, purpose, and expected arguments so a system can determine how to invoke it.
- How it is used
- Guides an LLM’s tool selection and argument formation by placing structured tool definitions in its context.
- How deep it goes
- Specifies five contract elements, from naming and parameter constraints to output and error schemas.
- Examples
- Uses a weather lookup and a bug investigation combining a function call, shell command, and MCP documentation.
- What you can do after
- Build a function-calling loop that sends a definition, executes the requested function, and returns its result to the model.
Also covered in
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- toolA software tool is a capability or utility that enables a person or system to perform a defined task or operation.
- JSON SchemaJSON Schema is a standard vocabulary and specification for describing, validating, and annotating the structure and constraints of JSON data.
- promptA prompt is text or other input supplied to an AI model to provide instructions, context, or other information that guides its response or generated output.
Related concepts
- software artifactkind of · A software artifact is a persistent work product or object produced, maintained, or consumed during software development or operation.
- tool useused by · Tool use is the capability of an AI system or agent to request external software operations through defined interfaces to obtain information or perform actions.
- toolconfigures · A software tool is a capability or utility that enables a person or system to perform a defined task or operation.
- docstringproduced by · A docstring is a string literal associated with a program element that documents its purpose, interface, or behavior for people or tools.
- agent harnessused by · An agent harness is the surrounding software, configuration, instructions, and execution environment that provides capabilities and manages an AI agent’s operation and constraints.
Questions about tool description
What is tool description?
A tool description is information specifying a callable tool’s purpose and how an agent or system can invoke it, including its inputs or parameters.
Which Manning books explain tool description?
Build an AI Agent (From Scratch), Vibe Engineering, AI Agents in Action, Second Edition and Context Engineering.
What should I know before learning tool description?
Manning books that teach tool description usually introduce language model, tool, JSON Schema and prompt first.
What is tool description a kind of?
Tool description is a kind of software artifact. A software artifact is a persistent work product or object produced, maintained, or consumed during software development or operation.
What are the parts of tool description?
Tool description includes actionable error contract.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.



