One format for every model and every stake.
Five production prompt formats, each born in a different lab and battle-tested in a different stack. Pick by model, by task shape, by stake. The Forge ships every output structured to one of these, so you ship to git, not to a chat window.
The Singapore GovTech framework that won the world's first GPT-4 prompting competition.
The procedural framework for tasks where the steps matter as much as the answer.
Anthropic's official Claude format — XML tags that lock in instruction adherence on long prompts.
OpenAI's official GPT-4.1 and GPT-5 production format — strict role-priority hierarchy.
From Google's 68-page Prompt Engineering whitepaper — the de-facto Vertex AI curriculum.
Not sure which one?
Drop your task in the Forge. We'll auto-pick the framework that fits the model, stake, and shape of your problem.
Open the Forge→