[BUG] Status bar shows "thinking with xhigh effort" while skill frontmatter sets effort: max

Resolved 💬 2 comments Opened May 20, 2026 by ivan01march Closed May 22, 2026

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

A skill with effort: max in its YAML frontmatter runs, and the status bar during the skill's thinking phase reads:

thinking with xhigh effort

The skill explicitly states it sets effort to max, and based on measurable signals (token usage, thinking duration, reasoning depth) I believe the runtime is in fact honoring effort: max — the frontmatter value is being applied. The bug appears to be confined to the thinking indicator label, which shows xhigh instead of max.

This looks like the same class of UI/runtime mismatch reported in #49538 (settings.json effortLevel enum excludes max, UI surfaces fall back to xhigh while the resolver correctly applies max to API requests) and #53217 (status bar / /effort slider disagree). The trigger here is different — skill frontmatter, not env var or settings.json — so it may be a partial clone rather than a full duplicate: the skill-frontmatter path likely reaches a different label-rendering branch that also doesn't know about max.

What Should Happen?

The thinking indicator should read thinking with max effort when the active skill's frontmatter sets effort: max, matching the effort level the runtime is actually applying.

Error Messages/Logs

Steps to Reproduce

  1. Create ~/.claude/skills/max-effort-test/SKILL.md:
   ---
   name: max-effort-test
   description: Minimal repro — skill that requests max effort via frontmatter.
   effort: max
   ---
   Answer the following, in order:

   1. What is 2+2? Show your reasoning step by step.
   2. Explain why 1+1=2 from Peano axioms.
   3. Explain Gödel's incompleteness theorem's implications for arithmetic.
  1. Start Claude Code on Opus 4.7 with any baseline effort (e.g. /effort xhigh).
  2. Invoke the skill: /max-effort-test
  3. Observe the status bar during the thinking phase.

Expected: thinking with max effort
Actual: thinking with xhigh effort

Claude Model

Opus

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

2.1.145

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

WSL (Windows Subsystem for Linux)

Additional Information

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗