Auto mode unavailable despite Max plan subscription

Status Closed — not planned
Reported on v2.1.97
Maintainer reply None cached
Activity 9 comments · opened Apr 9, 2026 · closed Apr 10, 2026

Bug Description

--mode auto flag is not recognized, and auto mode shows "not available for your plan" despite being on the Max plan with correct configuration.

Environment

  • Claude Code version: 2.1.97 (latest available via npm)
  • OS: macOS (Darwin 25.3.0)
  • Plan: Claude Max subscription

Steps to Reproduce

  1. Configure auto mode in ~/.claude/settings.json:

``json
{ "preferences": { "defaultMode": "auto" } }
``

  1. Launch Claude Code — auto mode is not activated, shows "not available for your plan"
  2. Try launching with claude --mode auto — returns error: unknown option '--mode'

Expected Behavior

Auto mode should be available for Max plan subscribers, and --mode should be a valid CLI flag.

Actual Behavior

  • Auto mode gated with "not available for your plan" message
  • --mode flag not recognized (CLI suggests --model instead)
  • Authentication refresh does not resolve the issue

Additional Context

  • ~/.claude/settings.json has defaultMode: "auto" correctly set
  • Confirmed version 2.1.97 is the latest available via npm show @anthropic-ai/claude-code version

View original on GitHub ↗

9 Comments

github-actions[bot] · 6 months ago

Found 3 possible duplicate issues:

  1. https://github.com/anthropics/claude-code/issues/33587
  2. https://github.com/anthropics/claude-code/issues/44708
  3. https://github.com/anthropics/claude-code/issues/42649

This issue will be automatically closed as a duplicate in 3 days.

  • If your issue is a duplicate, please close it and 👍 the existing issue instead
  • To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

ThatGuyInTheGlass · 6 months ago

👎 These are not duplicates of this issue.

The suggested issues (#33587, #44708, #42649) are about Plan mode breaking because the automode entitlement check bleeds into it. That's a different symptom.

This issue is about auto mode itself being completely unavailable on a Max subscription. I'm paying for Max. Auto mode is a Max feature. It doesn't work. Period. Shift+Tab doesn't cycle to it, --mode auto doesn't exist as a flag, and the settings config for automode is silently ignored.

Yes, the root cause is probably the same broken entitlement check — but closing this as a duplicate of Plan mode bugs means the actual auto mode unavailability gets buried and never explicitly tracked. Those issues are about Plan mode. This issue is about Auto mode. They're siblings, not duplicates.

The fact that there are now 4+ issues all pointing at the same busted permission gate should be a signal to actually prioritize fixing it, not to auto-close everything into a pile and forget about it.

Please reopen and track this separately, or at minimum confirm the entitlement check fix will address auto mode availability — not just the Plan mode regression.

CamdenClark · 6 months ago

same here

TioGlo · 6 months ago

It's pretty ridiculous. If we can use --dangerously-skip-permissions, auto mode needs to be accessible for MAX plan users and addressed independently of /plan mode.

yshrsmz · 6 months ago

I can enable auto mode with claude --permission-mode auto but claude code says "auto mode is unavailable for your plan" even though I'm on Max plan

timgu0 · 6 months ago
ThatGuyInTheGlass · 6 months ago

@timgu0 wow - totally not cool. firstly, may you please explain to me why it's not available for Max? :(
secondly, in the meantime, what's stopping us from simply manually adjusting the permissions to 'simulate' automode, in theory nothing right?

TioGlo · 6 months ago
Auto mode is not available for Max at the moment: https://code.claude.com/docs/en/permission-modes#eliminate-prompts-with-auto-mode

Exactly, we can use our MAX plan tokens to build the architecture to perform automode functionality. So lame. Why are y'all doing such a terrible job of supporting MAX plan users? Kissing the butts of massive corporate entities, while spitting in the faces of real people, then calling yourselves the good guys. What kind of alignment do you think you're really building with those business practices?

github-actions[bot] · 5 months ago

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.