many skills were written as elaborate itineraries or recipes. Models have gotten much better at understanding nuance and ambiguity, so overly specific guidance can now hinder results where it …
Cursor found that sections of their prompt that had been effective with earlier models needed tuning ... <maximize_context_understanding> Be THOROUGH ... they found it counterproductive with GPT-5 …
Removing repeated instructions and examples and simplifying tool descriptions can improve task performance and token efficiency. In a sample of internal coding-agent eval runs, configurations with …
Not every success criteria or failing eval is best solved by prompt engineering. For example, you can sometimes improve latency and cost more easily by selecting a different model.