DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.
The primary change allows individual prompt features to specify which Claude model they use. The pr-comments feature now explicitly uses claude-3-5-haiku-20241022 instead of the default model, optimizing for speed and cost-efficiency when processing pull request comments.
This update indicates a move toward task-specific model optimization, where different features can leverage the most appropriate Claude model based on their complexity and performance requirements.