What this error means

Unexpected value(s) prompt-caching-scope-2026-01-05 for the anthropic-beta header is a Claude Code failure pattern reported for developers trying to fix claude code sending prompt-caching-scope beta header to custom api endpoints that don't support it. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.

Why this happens

Claude Code extension v2.1.23 sends prompt-caching-scope-2026-01-05 beta header to custom API endpoints (via ANTHROPIC_BASE_URL) even when enablePromptCaching and useAnthropicBetaHeaders are both set to false. Results in API error on incompatible endpoints.

Common causes

  • When Claude Code is configured with ANTHROPIC_BASE_URL pointing to a custom endpoint (proxy, mock server, or alternative provider), it injects the prompt-caching-scope-2026-01-05 beta header even when prompt caching is disabled, causing 400 errors. Developers using custom API gateways encounter this incompatibility.
  • Claude Code extension v2.1.23 sends prompt-caching-scope-2026-01-05 beta header to custom API endpoints (via ANTHROPIC_BASE_URL) even when enablePromptCaching and useAnthropicBetaHeaders are both set to false. Results in API error on incompatible endpoints.

Quick fixes

  1. Confirm the exact error signature matches Unexpected value(s) prompt-caching-scope-2026-01-05 for the anthropic-beta header.
  2. Check the Claude Code account, local tool state, and provider configuration involved in the failing workflow.
  3. Compare the failing environment with a known working setup, then change one configuration value at a time.

Platform/tool-specific checks

  • Verify the command, editor, extension, or API client that produced the error.
  • Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
  • Avoid deleting credentials, local model data, or project settings until the failing scope is clear.

Step-by-step troubleshooting

  1. Capture the exact error message and the command, editor action, or request that triggered it.
  2. Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
  3. Review the source evidence below and compare it with your environment.
  4. Apply one change at a time and rerun the smallest failing action.
  5. Keep the working fix documented for the team or deployment environment.

How to prevent it

  • Keep provider/tool configuration documented.
  • Record non-secret diagnostics such as tool version, provider name, model name, and command path.
  • Add a lightweight check before CI or production workflows depend on the tool.