Crystal Sapphire Pokemon: Claude Code (Opus 4-7) vs. Codex (GPT 5.5)twitch.tv1 ポイント·投稿者 oelmgren·3 か月前·0 コメント
We Built a Claude Agent That Doesn't Know Its Own API Keyslistenlabs.ai8 ポイント·投稿者 oelmgren·4 か月前·2 コメント
oelmgren·3 か月前·議論Is there evidence that they use that information for anything other than browser fingerprinting or fraud detection?That seems like the most obvious use case? Or maybe I missed something in the write up.
oelmgren·4 か月前·議論This is pretty cool, have personally felt this limitation many a time.Basically been relying on spinning up cursor / niteshift / devin workflows since they have their own containers but this could be interesting to keep it all on your main machine.
oelmgren·4 か月前·議論Yeah, this is exactly what I was thinking.Maybe they’ll add the ability to sandbox the env variables from the agent’s bash execution via the Agent SDK settings at some point.
That seems like the most obvious use case? Or maybe I missed something in the write up.