Cache is invalidated increasing costs

Summary

Luna 5.6 xhigh.
I've sent one prompt and agent was implementing it for 9 minutes, sending about 60 requests, quite evenly so input cache could not expire (OpenAI has 30 minutes retention). Context was steadily growing from 25K to 135K. There was no context compaction. Despite pure agentic work cache was invalidated 3 times during these 9 minutes, causing substantial costs increase. Here are traces id that causes new full cache write for input.
Traces ID:
688515c1ff04db92149a370635b6dfbc
f591b3c3861e52523c99049e28542125
80f39ae574ac6b340530bfe244d01a6a

Seems there is some tool call that is modifying prior context and causing full cache invalidation

Expected Behavior

As long as there is no context compaction or cache time has not expired there should not be cache invalidation with full context cache write.

Actual Behavior

Cache is invalidated when it should not.

Steps to reproduce the issue

Hard to tell what is causing it. Providing traces id:

1. before invalidating cache: a9ca8bb342da0e7d4fcca5498a25bf0e
   next request invalidating cache: 688515c1ff04db92149a370635b6dfbc
2. before invalidating cache: ca3e9dfbf07552b61e0b5c486d0ffd06
   next request invalidating cache: f591b3c3861e52523c99049e28542125
3. before invalidating cache: 99eb39484857d759ba5d9ad7e999f837
   next request invalidating cache: 80f39ae574ac6b340530bfe244d01a6a

Command Code Version

1.11.0

Operating System

Linux

Terminal/IDE

JetBrains-JediTerm

Shell

zsh

Session file (optional)

No response

Fix prompt (optional)

No response

Additional context

No response

Bắt đầu bằng cách so sánh các cặp trace được liệt kê xung quanh mỗi lần vô hiệu hóa cache, sử dụng Command Code 1.11.0 trên Linux với hành vi caching của OpenAI đã được báo cáo. Xác định request hoặc lệnh gọi công cụ nào thay đổi ngữ cảnh trước đó dù không có quá trình nén hay hết hạn; được xem là hoàn tất khi các request mang tính agent lặp lại vẫn giữ input cache trong những điều kiện đó và tránh các lần ghi toàn bộ cache.