BriefPulse Automation · Automation reporting with the failure modes left in. RSS · BriefPulse network
BriefPulse Automation

Workflows, tools and operating patterns that keep working after the demo.

16 September 2026

Brief

Claude Code v2.1.267 adds a provider-wide effort cap and a fresh-prompt switch

A point release adds a maxEffortLevel setting that caps effort across providers, plus a flag to stop reusing a conversation's recorded system prompt. It also bundles a batch of reliability fixes.

What this story rests on:  4 verified figures · 2 sources cited
Original graphic. Every figure in it is stated in the reporting; the sources are listed below this article.

The release adds a maxEffortLevel setting, which can sit at the top level or per model under modelSettings. It caps the effort level on every provider, including Bedrock, Vertex and Foundry, and users can still pick a lower level than the cap.

For anyone iterating on prompt text, --system-prompt-snapshot off renders the system prompt fresh on every request instead of reusing the conversation's recorded prompt.

The rest is repair work: scheduled cloud tasks failing at startup for organizations whose managed settings require sandboxing, blank local command output on mobile clients, and artifact publishes cut off by a dropped connection mid-upload, which now retry once when Claude Code can tell the upload never completed.

Our reading

The effort cap is the piece most likely to matter operationally, because it gives an operator one place to bound cost or latency across providers rather than tuning each one.

Source details and supporting facts

Each line is stated by the page named above it.

Stated by github.com

  • Added maxEffortLevel setting (top-level or per model under modelSettings): caps the effort level on every provider, including Bedrock, Vertex and Foundry; users can still pick a lower level
  • Added --system-prompt-snapshot off to render the system prompt fresh on every request instead of reusing the conversation's recorded prompt (for iterating on prompt text)
  • Fixed Cowork scheduled tasks in the cloud failing at startup for organizations whose managed settings require sandboxing
  • Fixed artifact publishes cut off by a dropped connection mid-upload: they now retry once when Claude Code can tell the upload never completed

Sources

  1. Anthropic Claude Code release notes (GitHub)Text stored 15 September 2026

How this story was checked. Written from the 1 page listed above, stored 15 September 2026; claims checked against that stored text on 15 September 2026.

What that means
  • 4 of 4 reported statements were confirmed against the page that carries them; the rest were removed rather than published.
  • Figures in the text were required to appear in the stored source text: yes. Identifiers: yes.
  • The check reads stored text only: no claim rests on a fresh look that did not happen.
  • Where the reporting was silent, the text says so instead of filling the gap.

More from Automation