ClaudeDevs

To learn more, see our docs on automatic-caching: #automatic-caching" target="_blank" rel="nofollow noopener">platform.claude.com/docs/en/build-… And see our docs on mid-conversation system prompt updates: platform.claude.com/docs/en/build-…

With Opus 4.8, you can pass a system-role message mid-conversation. Claude treats this as an authoritative instruction from that point in the conversation onward. Pair this auto-caching to ensure that these updates are added to the prompt cache. x.com/RLanceMartin/s…

With Opus 4.8, you can add system instructions mid-conversation without breaking the prompt cache. More cache hits means lower cost and latency for your API requests.

Available today on Max, Team, Enterprise, and via the API — including Bedrock, Vertex AI, and Foundry. On by default for Max and Team plans, Enterprise admins can opt in via managed settings. Docs: code.claude.com/docs/en/workfl…

More on how the orchestration works, and what early users have built with it, on the blog: claude.com/blog/introduci…

New in Claude Code (research preview): dynamic workflows. Claude writes an orchestration script on the fly, then spins up a large fleet of coordinated subagents in parallel to take on your most complex tasks. Use the word "workflow" in a prompt to get started.

See our prompting guide for more tips and best practices for working with Opus 4.8: platform.claude.com/docs/en/build-…

We’ve updated Claude Code's built-in claude-api skill migration guidance for 4.8. Run "/claude-api migrate" to update your model strings and suggest prompt improvements that are tuned for Opus 4.8.

Opus 4.8 is live in Claude Code today. A few things worth knowing: 🧵

And we’re still working! Please let us know if there are any bugs that are impacting your experience.

It’s now easier than ever to send feedback. When you hit /feedback, you can now just send over the last day or week of sessions instead of having to remember which session had the bug in it.

We’ve been putting a lot of effort into making Claude Code more responsive & reliable. Here’s an update on everything we’ve done: