Prompt Template
定义
A prompt template is a reusable set of instructions for a large language model with placeholder variables that are filled at runtime with specific data. Unlike one-off prompts typed into a chat window, templates are versioned, tested, and optimized over time. In JieGou, every AI recipe is built on a prompt template with typed input schemas, output schemas, and optional knowledge base context.
Templates vs. Raw Prompts
A raw prompt is disposable — you type it once and get one response. A template is an asset — it's versioned, tested across multiple inputs, evaluated with AI Bakeoffs, and improved over time. Templates separate the "what to do" (the prompt logic) from the "what to do it with" (the input data), enabling reuse across hundreds of runs.
Variables and Schemas
Templates use {{variable}} syntax for input placeholders. Each variable is defined in an input schema with a type, description, and validation rules. When a recipe runs, JieGou validates the input against the schema, substitutes variables into the template, and sends the resolved prompt to the LLM. This ensures every execution receives well-formed instructions.