Sign in
← AI PulseSep 9, 2026

Wire · news · Single-source brief

Claude Code Updates Include maxEffortLevel and System Prompt Snapshot Control

Claude Code has introduced a new maxEffortLevel setting to cap effort across providers and a system-prompt-snapshot option for prompt iteration.

By Illumora Editorial

Source · Sep 9, 2026, 7:58 PM · On Illumora · Sep 9, 2026, 8:02 PM

Media from the primary source — shown here so you can stay on Illumora.

Rewritten from one allowlisted primary — not independent enterprise reporting. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →Claude Code What's New + releases — Claude Code changelog - Claude Code Docs
Save

Claude Code has implemented several updates, including a new setting to manage effort levels and a feature for prompt iteration. These changes affect how users interact with various providers and refine their prompts.

Key Points

  • A new maxEffortLevel setting has been added, available at the top-level or per model under modelSettings.
  • This maxEffortLevel caps the effort level on every provider, including Bedrock, Vertex, and Foundry.
  • Users retain the ability to select a lower effort level than the maximum set.
  • The --system-prompt-snapshot off option allows rendering the system prompt fresh on every request.
  • This option is designed for iterating on prompt text, preventing reuse of the conversation's recorded prompt.
  • Several fixes address issues such as Cowork scheduled tasks failing at startup for sandboxed organizations and blank output rendering on mobile clients.
  • Other fixes include addressing shift+enter and option+backspace not working after reconnecting to tmux or ssh sessions in an agent view.

Context

According to the Claude Code changelog, these updates introduce new controls for managing model interaction and prompt development. The changes also address various bug fixes across different platforms and functionalities, enhancing the stability and user experience of Claude Code.

Why It Matters

These updates provide builders with more granular control over model resource utilization and prompt engineering workflows. The maxEffortLevel setting can help manage costs and performance, while the system prompt snapshot control streamlines the iterative process of prompt refinement.

What To Do

  • Review the maxEffortLevel setting to understand how it can cap effort across providers like Bedrock, Vertex, and Foundry.
  • Test the --system-prompt-snapshot off option when iterating on prompt text to ensure fresh prompt rendering on each request.
  • Note the various bug fixes, especially if using Cowork scheduled tasks, mobile clients, or tmux/ssh sessions within an agent view.
  • Check for updated documentation regarding these new settings and features within Claude Code.

Keep Exploring

/atlas/claude-family /techniques/system-user-separation