Claude Runs Pointless 24-Hour Optimization, Burning $800 In Compute
Source: https://github.com/chatgptprojects/clear-code/blob/642c7f944bbe5f7e57c05d756ab7fa7c9c5035cc/src/utils/userPromptKeywords.ts#L8
Perhaps this can be the basis for a tool that helps people work with Claude to make sure they get their message across to the team: import re claude_code_regex = re.compile(r"\b(wtf|wth|ffs|omfg|shit(ty|tiest)?|dumbass|horrible|awful|piss(ed|ing)? off|piece of (shit|crap|junk)|what the (fuck|hell)|fucking? (broken|useless|terrible|awful|horrible)|fuck you|screw (this|you)|so frustrating|this sucks|damn it)\b", re.I)
Perhaps this can be the basis for a tool that helps people work with Claude to make sure they get their message across to the team:
import re
claude_code_regex = re.compile(r"\b(wtf|wth|ffs|omfg|shit(ty|tiest)?|dumbass|horrible|awful|piss(ed|ing)? off|piece of (shit|crap|junk)|what the (fuck|hell)|fucking? (broken|useless|terrible|awful|horrible)|fuck you|screw (this|you)|so frustrating|this sucks|damn it)\b", re.I)
I'm concerned that your response won't trigger the Claude Code `negativePattern`. All-caps is okay because they downcase, but you need a straight-up "f you" – but I believe you have to actually drop the F-bomb, or add "ffs" somewhere. Otherwise this risks looking like a perfectly successful interaction.
I have worked on swears for over 25 years, and I would be happy to donate an improved regex to the Claude Code team. I feel like this could help make everyone more successful, and then I would have a good answer when people ask me what the point of working on swears is.
Source: https://github.com/chatgptprojects/clear-code/blob/642c7f944bbe5f7e57c05d756ab7fa7c9c5035cc/src/utils/userPromptKeywords.ts#L8