A new tool from Anthropic to improve your prompts by following best practices for more reliable output.
Take your existing prompt and automatically refine it. It will optimise the prompt for use with Claude. The prompt improver uses methods such as Chain-of-thought reasoning, will improve structure, format examples correctly in XML format that Claude prefers, add additional examples and will prefill the assistant message to help enforce output formats.
They claim improved accuracy by 30% for classification tasks, plus an adherence to an instruction to write summaries within a specific word count range of 100%!
They say adding examples it one of the most effective ways to improve the model output quality. So this new tool can help generate examples which overall improves accuracy of understanding instructions, consistent output formatting and performance in handling complex tasks.
Their other tools include prompt evaluator to help benchmark and improve performance, so you can test the “improved” prompt.