[MODEL] opus 4.6

Preflight Checklist

• [x] I have searched existing issues for similar behavior reports
• [x] This report does NOT contain sensitive information (API keys, passwords, etc.)

Type of Behavior Issue

Claude ignored my instructions or configuration

What You Asked Claude to Do

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking
A question and ordering compliance
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care

4.5 was perfect in comparison

What Claude Actually Did

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking
A question and ordering compliance
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care

4.5 was perfect in comparison

Expected Behavior

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking
A question and ordering compliance
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care

4.5 was perfect in comparison

Files Affected

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking 
A question and ordering compliance 
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere 

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care 

4.5 was perfect in comparison

Permission Mode

Accept Edits was ON (auto-accepting changes)

Can You Reproduce This?

Yes, every time with the same prompt

Steps to Reproduce

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking
A question and ordering compliance
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care

4.5 was perfect in comparison

Claude Model

Opus

Relevant Conversation

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking 
A question and ordering compliance 
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere 

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care 

4.5 was perfect in comparison

Impact

Critical - Data loss or corrupted project

Claude Code Version

Claude code/Opus 4.6

Platform

Anthropic API

Additional Context

Was LLM in the loop completely removed in opus 4.6 Claude code because it's severely lacking context interpretation during orchestration in Claude code, also reflected in your comparison chart 4.5 vs 4.6 where 4.5 is ahead of 4.6
It's simply completely unable to connect context information, as if the orchestration team was on holidays when you released 4.6

It's like asking
A question and ordering compliance
Referring to it and claude code simply hallucinating as if nothing was ever looped from previous context, did you forget to include the cached entries somewhere

Also during session resets the model runs Amok without reading previous context and nobody in antropic seems to care

4.5 was perfect in comparison