Anthropic urges users to update prompts for Claude Opus 5.5
Anthropic says users should abandon several prompting practices developed for older model generations when working with Claude Opus 5.5. The model is now the default on the Pro, Max, Ultra, and Team paid plans. The company says phrases such as “think hard” and “think carefully” are unnecessary because Opus 5.5 reasons automatically and adjusts its depth to the task and selected effort level. In Anthropic’s tests, removing those phrases made responses begin faster without a measurable quality loss.
The company recommends starting with the default Medium effort setting. Anthropic says Opus 5.5 at Medium matched the performance of Opus 5 at High in internal tests, while higher settings cost more and do not always improve results. High can be selected when Medium is insufficient.
Anthropic also advises giving the goal, output format, and conditions for follow-up questions in one initial request. Users are encouraged to attach diagrams and tables, request finished documents directly, specify unwanted design styles precisely, and connect Claude to relevant files and systems. The company says Opus 5.5 is better at completing multi-stage tasks and reading dense charts than its predecessor.
Written by Sakhanda Wire AI from reporting by The Decoder. Facts, quotes and full context are in the original article below.