How to use it
- Add a system instruction, context, and the task.
- Choose the output role.
- Click Build prompt and copy the result.
Why use it
- Structure prompts for more consistent outputs.
- Separate instructions from context and task.
- Roughly estimate token cost before you run a prompt.
- Keep your prompts private; nothing is uploaded.
- Free and reusable for any project.
Why structure your prompts?
Large language models respond better when instructions, context, and the task are clearly separated. A wrapped-up blob of text is harder for a model to interpret than a prompt that explicitly labels the system's role, the background information, and the user's request. Structuring your prompts this way leads to more reliable, on-topic responses.
Common prompt sections
- System: sets the model's role, tone, and constraints (for example, "You are a technical writer who explains simply").
- Context: the background or data the model needs to answer correctly.
- Task: the specific instruction: what to produce, in what format, for whom.
Tips for better prompts
- Be specific about the desired output format and length.
- Provide an example whenever possible.
- Break complex requests into smaller, single-purpose prompts.
- Iterate: small changes to wording can change results significantly.
The builder assembles these sections into a clean, copy-ready prompt and estimates its size in tokens, so you know roughly what a run will cost in the AI cost calculator.