Skip to content
CC Switch WikiUnofficial Community Wiki

temporarily unavailable, so auto mode cannot - Claude Code

Quick fix

Upgrade cc-switch to v3.15.0+ for the model mapping and thinking conversion fixes, or override the default model env var in general config.

Symptom

Section titled “Symptom”
Error output
Error: claude-opus-4-7[1m] is temporarily unavailable, so auto mode cannot determine the safety of Bash right now.

Cause

Section titled “Cause”

Several causes converge here. First, the Auto Mode classifier uses specific model names (claude-sonnet-4-6 or claude-opus-4-7[1m]) for its safety judgement, and early versions may not have routed those names correctly. Second, under api_format=openai_chat, `"thinking": { "type": "disabled" }` in the request was not converted into the upstream thinking-off parameter (enable_thinking: false), so upstream kept emitting reasoning and exhausted max_tokens.

Environment variable and parameter compatibility also trigger it. The default model ID cc-switch writes into settings.json may not match the third-party backend (GitHub Copilot, for example), so the fallback resolution fails; and some upstream APIs do not support the stop_sequences parameter the classifier sends, returning a 400.

Fix

Section titled “Fix”

  1. Upgrade cc-switch to v3.15.0 or newer. That version fixed keyword substring mapping for classifier model names, and thinking.type=disabled not being converted correctly under openai_chat, which exhausted tokens.

  2. If it still errors on a backend such as GitHub Copilot, override the default model environment variable by hand in cc-switch Common Config, matching exactly a model ID the backend supports.

    // settings.json
    {
      "env": {
        "ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-opus-4.6",
        "ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-opus-4.6",
        "ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4.6",
        "ANTHROPIC_MODEL": "claude-opus-4.6",
        "ANTHROPIC_REASONING_MODEL": "claude-opus-4.6"
      }
    }

  3. If you confirm the 400 comes from upstream (Copilot) not supporting stop_sequences, wait for a later cc-switch version to strip the stop parameter on that path, or avoid that backend under Auto Mode for now.

Affected Versions

Section titled “Affected Versions”
ToolClaude Code
Version2.1.90 - 2.1.112
PlatformsWindows

Source Issues

Section titled “Source Issues”

This page is distilled from 5 real issues

FAQ

Section titled “FAQ”
Why do ordinary conversations work while only Auto Mode errors?
The Auto Mode classifier uses specific model names — ones with a [1m] suffix, for example — for its safety judgement. If cc-switch does not map those names correctly, or does not convert thinking: disabled correctly under openai_chat, the classifier request fails.
It still errors with GitHub Copilot after upgrading. What now?
The default model ID cc-switch writes may not match the models available on Copilot, or the Copilot API may not support stop_sequences. Try overriding the ANTHROPIC_DEFAULT_*_MODEL environment variables by hand in general config, or wait for a later fix.
Section titled “Related problems”
settings.json Full Overwrite Causes Plugin FailureUpgrade to v3.20.1+ to fix the bug that clears enabledPlugins when switching configs.9 issues mergedAPI Error: 400 Param Incorrect Claude CodeUpgrade CC Switch to v3.16.2+ to fix 400 errors caused by missing reasoning_content in MiMo thinking mode.2 issues mergedLow cache hit rate around 40% and very slow responses in Claude CodeSet CLAUDE_CODE_ATTRIBUTION_HEADER to false in .claude/settings.json to stop dynamic headers from breaking third-party model prompt caching.1 issues merged

This is an unofficial community wiki with no affiliation to the cc-switch authors or the project itself. Its content is compiled from the project's public GitHub issues. This site distributes no software.