It felt unreasonable when I was using it at midnight and had to wait 10 minutes to refill the KV cache on my local GPU :-)

A simple solution here would be to evaluate the date once per session, or once each time the `opencode` binary is launched (to avoid old, long-running sessions getting stuck in the past).

Or they could add the frozen initial starting date once at the start, then inject an additional "the date is now X" message any time midnight passes.

But then someone would write an article complaining about how the newly-injected date confused their agent!

I was confused about how this was written in the article. Sure, include system time, or a simple get current time tool, whatever, it’s inconsequential for the most part, but who is out here changing datetimes in a conversation and invalidating prefixes?

If you need to, just prefix user messages with the current datetime.

Why is this even a complaint?

I suspect that if you prefixed every user message with a date the model might get confused and start treating the date as more relevant to the current task than it actually is.