Anthropic prompt caching fails when there are more than 4 cache points #3467
Description
Initial Checks
- I confirm that I'm using the latest version of Pydantic AI
- I confirm that I searched for my issue in https://github.com/pydantic/pydantic-ai/issues before opening this issue
Description
As pointed out in #3453.
We should post-process the messages sent to the Anthropic API to strip all cache markers but the last 4.
Example Code
Python, Pydantic AI & LLM client version
latest