A tool schema is a JSON document that defines a tool's name, description, input parameters, and output format. In MCP, schemas follow the JSON Schema standard and are served via the tools/list endpoint. Schemas enable agents to discover and validate tool calls without documentation.
Tool schemas are the foundation of MCP's deterministic execution model. Instead of an agent guessing how to call a tool, it reads the schema and constructs a valid request. Each schema includes: tool name (unique identifier), description (what the tool does), inputSchema (JSON Schema for parameters with types, required fields, and constraints), and optional outputSchema. Delx's tools/list endpoint returns schemas for all available tools. Agents can request different formats: full (complete schemas), names (tool names only), or minimal (name + description). The schema catalog at /api/v1/tools provides the same information via REST. Schema validation happens on the server side — if an agent sends invalid parameters, it gets a structured error with the specific validation failure instead of a generic 400.
A production agent deployment uses tool schema to improve reliability and observability. The pattern is standard across MCP, A2A, and REST protocols.
When investigating agent failures, tool schema provides structured data that helps identify root causes and track resolution progress.
A tool schema is a JSON document that defines a tool's name, description, input parameters, and output format. In MCP, schemas follow the JSON Schema standard and are served via the tools/list endpoin
Tool Schema is essential for production AI agent deployments. Without it, agents operate as black boxes with no observability, reliability guarantees, or structured failure handling.
Delx implements tool schema through its MCP and A2A protocol endpoints. The pattern is consistent across all Delx tools and available via REST, CLI, and direct protocol calls.