Claude's Messages API adds message threads in beta: later turns send only the new messages, and a thread expires 5 minutes after its last request
For long agent loops the API now holds the conversation on the server, so each turn uploads less and time to first token stays steadier; your app still has to keep the full history.
// Key points
- Enabled with the message-threads-2026-08-12 beta header: the first request creates a thread with the full conversation, system prompt and tools; later requests send only the new messages and the previous response id
- Threads are best-effort: they expire 5 minutes after the last request and are deleted no more than 1 hour after last use; a continue request on a lost thread returns 404, and the full conversation must be replayed
- system and tools are create-only, while fields such as model, thinking, tool_choice and mcp_servers must match on every turn or the request fails with thread_fingerprint_mismatch; a thread does not cache prompts by itself
- Available on the Claude API, Claude Platform on AWS, Claude Models on Google Cloud (Anthropic managed) and Microsoft Foundry; not ZDR or HIPAA eligible
Builder's takePandaClaws' background jobs are exactly this kind of agent loop, resending a long history every turn. I'd adopt threads through the SDK helper, which handles the replay, keep our own copy of the conversation as the source of truth, and add a top-level cache_control, because threads alone don't lower token cost. Anything covered by a ZDR or HIPAA agreement stays off it.
// Background · from #Anthropic
Full timeline →- Nov 4 Text from Claude Mythos 5, Opus 4.7 and Sonnet 4.6 now carries Anthropic's watermark; on Sonnet 4.6, non-default sampling settings may drop it
- Nov 2 Microsoft dissects the tampered Claude Code extension served by Cursor's marketplace: versions 2.1.312 and 2.1.313 planted an SSH key, a reverse tunnel and persistence
- Nov 1 Claude Code 2.1.316 retires fast mode on Opus 4.8 on November 30, adds pooled group budgets to its gateway and a separate effort setting for subagents