Schema

A formal definition of a data structure: what fields exist, what types they are, which are required, and what values are valid. In the AI tooling context, schemas define tool parameters (JSON Schema), API request/response formats (OpenAPI), and data models. MCP tool definitions include parameter schemas that tell AI agents exactly what inputs a tool expects.

If a database is a spreadsheet, a schema is the column headers, data types, and validation rules, except enforced by software instead of a stern email from your project manager.

Why it matters for writers: Schemas are one of the most documentation-dense artifacts in software development. A well-documented schema reduces integration errors, improves developer experience, and (in the MCP context) helps AI agents use tools correctly. Technical writers who can read and explain schemas are increasingly valuable as AI tool ecosystems grow. It's a superpower that doesn't require learning how to train a model. Just how to read a JSON file and explain it like a human.

Related terms: Model Context Protocol · Tool Use · MCP Server