XML tags

XML tags are a prompting technique where you wrap different sections of your prompt in descriptive markup elements—such as <instructions>, <context>, <document>, or <input>—to help Claude clearly distinguish between instructions, reference data, examples, and user content. By giving each type of content its own named container, you eliminate ambiguity about what Claude should treat as a command versus background information, reducing misinterpretation especially in long or complex prompts. This is not a special model feature or API setting you toggle on—it is simply standard XML syntax written directly into your system prompt or user messages. Claude has been trained to recognize and weigh these structural cues natively, so no additional configuration is required. The technique works across all Claude models and all subscription plans. The value of XML tags scales with prompt complexity. A short, simple prompt rarely needs them. But once your prompt combines instructions, multiple documents, few-shot examples, and variable user input, XML tags prevent 'context bleeding'—the failure mode where Claude cannot tell where one section ends and another begins—and make responses more consistent and easier to parse downstream.