This PR doesn't have any linked issues. Please open an issue that references this PR. From there we can discuss and prioritise.

It's a little annoying that we have to probe process like that. Is there any other signal?

what kind of an automated signal could we probe? I mean - is there even any other way to accomplish this than to probe a global (either directly checking something on process, or similar, or process.env)? I think that the Replay team could be open to introducing something else - but I truly don't know what else can be used to signal this.

I was hoping for an environment variable 😅

Is there any particular reason why you are more OK with env var than with a process property? For context - I originally asked them if there is any env var that could be used to detect this because that was intuitively what I was expecting to be exposed. That's how I've learned that there is this process property available.

So my intuition was kinda like yours - but I don't mind any of those ways to detect something. It seems that this global object (process.recordreplay) has some additional methods exposed on it (like you can get a predicted ID of the recording, the current "pause point" and stuff like that). So from their PoV, it makes sense that this is an object on the process as they wouldn't be able to expose methods using env vars alone. So exposing additional env var could be seen as somewhat redundant from their PoV here as this global object will exist anyway.

If you are strongly opposed to checking this global object then I can circle back to the Replay team and ask if there is any chance that they would add some env var but I wonder if that's truly necessary.