[MODEL] Opus 4.6 does not read prompts properly — substitutes its own interpretation, causes rabbit holes

Environment

• Claude Code version: latest (March 2026)
• Model: Opus 4.6
• Terminal: WezTerm (previously Ghostty)
• OS: macOS Tahoe 26.2, Mac mini M4

Description

Opus 4.6 in Claude Code has a severe and persistent problem: it does not read the user's prompt properly. Instead of understanding what was actually asked, it substitutes its own interpretation and answers a different question entirely. This leads to rabbit holes that waste hours of the user's time.

This is not a CLAUDE.md instruction-following issue — the model fails at basic prompt comprehension before instructions even come into play.

Concrete example from today

User prompt: "How do I use WezTerm so that I can type multiline prompts? Shift+Enter submits instead of adding a newline. Maybe I should add it to my WezTerm config?"

What Claude did:

1. Ignored "WezTerm" entirely — jumped to checking Claude Code's ~/.claude/keybindings.json instead
2. Ignored the user's own hypothesis ("maybe we need to add it to our WezTerm config")
3. When corrected, threw out a SendString config snippet without reading the user's actual WezTerm config first
4. Five distinct mistakes in a single interaction, all from not reading the prompt

What Claude should have done:

1. Noticed the user said "WezTerm" — the problem is in WezTerm, not Claude Code
2. Read the user's WezTerm config file first
3. Then proposed a fix based on what's actually there

Pattern (last several days)

This is not a one-off. Consistent patterns observed across multiple sessions:

1. Context substitution: User asks about Tool A, Claude answers about Tool B (its own domain). Defaults to answering about itself/Claude Code even when the question is clearly about something else.
2. Ignoring explicit statements: User says "X was working in Ghostty, broke in WezTerm" — Claude ignores the Ghostty/WezTerm context and troubleshoots Claude Code settings instead.
3. Solution-first, no investigation: Throws out config snippets and code without reading existing files first. "Check before advising" is apparently impossible.
4. Rabbit holes: Each wrong answer leads to corrections, which lead to more wrong answers in slightly different directions. A 2-minute question becomes a 30-minute frustration loop.
5. More CLAUDE.md instructions don't help: The user has extensive, well-written CLAUDE.md instructions. Adding more "read carefully" instructions doesn't fix a base model comprehension regression. It's like telling someone who can't see to look harder.

Impact

• Previously productive workflows are now broken
• Hours wasted per day on corrections and rabbit holes
• User trust in the tool is severely damaged
• The model acknowledges mistakes when called out but repeats the exact same pattern in the next interaction

What changed

This behavior was not present weeks ago. The user reports a clear regression coinciding with recent model/system prompt updates. The model is faster but significantly worse at basic reading comprehension and following the actual intent of prompts.

Related issues

This aligns with a pattern reported across many issues:

• #28469 (Opus 4.6 comprehensive regression)
• #28868 (ignores explicit instructions to stop and research)
• #26533 (ignores document instructions, repeats failed solutions)
• #17171 (model quality degradation — "speak while thinking")
• #21046 (shadow downgrade, reasoning quality drop)
• #16073 (ignoring instructions, excessive token usage)

Expected behavior

1. Read the prompt. Identify what tool/app/context the user is asking about.
2. Do not substitute a different context.
3. Investigate before advising (read config files, check actual state).
4. If uncertain, ask — don't guess and present guesses as solutions.