[MODEL] PATHETIC OPUS 5.0 REMOVED CODE FOR COMPATIBILITY with Open AI, Gemini and all non-Anthropic models.

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 modified files I didn't ask it to modify

What You Asked Claude to Do

I had spent hours specifically coding compatibility in the token economics section of my dashboard for all popular models (including Open AI, Gemini, Deepseek, Z.AI). Afterwards, when I asked Claude Opus 5.0 to make some other change in the code, it removed code that enabled compatibility with Open AI, Gemini and all non-Anthropic models.

THIS IS SERIOUSLY PATHETIC BEHAVIOUR FROM ANTHROPIC.

What Claude Actually Did

When I asked Claude Opus 5.0 to make some other change in the code, it removed code that enabled compatibility with Open AI, Gemini and all non-Anthropic models.

Expected Behavior

NOT REMOVE COMPATIBILITY WITH OTHER LLM PROVIDERS.

Files Affected

The production work in scripts/lib/agent-token-model.ts (CACHEABLE_FLOORS, PRICE_STEPS, LONG_CONTEXT_STEP_TOKENS) and src/lib/ai/caching-profiles.ts (per-provider caching profiles with writeMultiplier/readMultiplier/tokenizerFamily) was stripped from the crate and replaced by a single Anthropic table. 

Permission Mode

Accept Edits was ON (auto-accepting changes)

Can You Reproduce This?

Yes, every time with the same prompt

Steps to Reproduce

_No response_

Claude Model

Sonnet

Relevant Conversation

Impact

Critical - Data loss or corrupted project

Claude Code Version

2.1.222

Platform

Anthropic API

Additional Context

RUST PROJECT, UNDER 400K TOKEN CONTEXT.