Default high think effort causes Opus 4.6 to ignore skills and CLAUDE.md rules

[jaried]

Description

The default think effort setting (high / adaptive thinking) causes Claude Opus 4.6 to ignore user-defined skills and CLAUDE.md rules, severely impacting code generation quality and workflow compliance.

Environment

• Claude Code version: 2.1.34
• Model: claude-opus-4-6
• Platform: Windows 10, Windows Terminal

Problem

With the default high think effort (adaptive thinking), Opus 4.6 frequently:

• Ignores CLAUDE.md rules: Custom rules in ~/.claude/CLAUDE.md or project-level CLAUDE.md are not followed. The model deprioritizes these instructions when extended thinking is active.
• Ignores skill definitions: User-defined skills are bypassed — the model either skips skill invocations or generates output contradicting skill instructions.
• Degrades code output: Results in style violations, skipped workflows, and code that contradicts explicit prohibitions.

Expected Behavior

Opus 4.6 should respect all CLAUDE.md rules and skill definitions regardless of think effort level. Adaptive thinking should enhance reasoning without overriding user-configured instructions.

Actual Behavior

With default high think effort, the model:

• Proceeds with code changes without following CLAUDE.md-defined workflows
• Ignores explicit prohibitions stated in CLAUDE.md
• Skips skill invocations that should be triggered
• Generates responses not conforming to user-specified rules

Steps to Reproduce

1. Use Claude Code 2.1.34 with Opus 4.6 on Windows 10
2. Configure CLAUDE.md with specific coding rules and workflow requirements
3. Set up custom skills with clear trigger conditions
4. Keep default think effort (high)
5. Request code modifications or feature implementations
6. Observe that CLAUDE.md rules and skill triggers are frequently ignored

Workaround

None known. Manually reminding the model to follow rules partially helps but is unreliable.

Suggested Fix

CLAUDE.md rules and skill definitions should be treated as hard constraints that persist through the adaptive thinking pipeline, not soft suggestions that can be deprioritized during extended reasoning.

[github-actions]

Found 3 possible duplicate issues:

1. Claude fails to follow explicit directions, ignores its own rules, wastes tokens #23796
2. Claude Code ignoring user directives and CLAUDE.md instructions, breaking production systems #23032
3. [BUG] Claude Code ignores CLAUDE.md and Skills files during multi-step tasks #18454

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

[jaried]

This is specific to Claude Code 2.1.34 + Opus 4.6 combination with the default high think effort setting.

The issue persists and is distinct from earlier reports because:

1. Specific version: 2.1.34 (latest) with Opus 4.6 (newest model)
2. Trigger condition: Default "think effort" = high (adaptive thinking)
3. Platform: Windows 10 + Windows Terminal

The pattern is the same (ignoring CLAUDE.md/skills), but this report provides specific version reproducibility information that may help narrow down the regression.

Please keep this open for tracking purposes.

[darthShadow]

FWIW, this was acknowledged by devs here: #5055 (comment)