βThis meeting could have been an email, and somehow the email would still ask for a follow-up.β
Generate office-safe roasts for meetings, coworkers, reply-all chaos, and corporate banter without explicit or hateful content. Practice them in PG mode.
What did they say? Get your comeback in seconds. Free, no signup.
Aim at behaviors, not identity. Roasting a calendar invite, an overused buzzword, or a reply-all habit is safer and funnier than attacking a person. The rule of thumb: if the line would still work if a different coworker did the same thing, it is about the behavior. If it only works because of who they are, leave it out. HR-safe and funny are not opposites; specificity does both jobs.
Everyone knows the meeting that should have been an email, the deck with 74 slides, and the coworker who says circle back three times before lunch. Recognition is the engine of office comedy. A roast about a shared pain point gets a laugh from the whole channel, while an inside joke about one person gets an awkward silence and a message from your manager. Start from what everyone has suffered through.
The office-safe style in the comeback generator sounds like a performance review that learned sarcasm, which makes it perfect for workplace satire. Generate a line, read it out loud, and ask whether it would survive being screenshotted. Then take it into a PG-mode battle: the AI judge rewards clean wordplay just as much as savage material, so practicing clean does not mean practicing weak.
βThis meeting could have been an email, and somehow the email would still ask for a follow-up.β
βYour reply-all had more distribution than strategy, which is impressive in the worst possible way.β
βThat deck has so many slides it qualifies as a remote work location.β
βYou scheduled a thirty-minute sync to decide when to schedule the sync.β
βYou have said synergy so many times it is now a line item in the budget.β
βYour camera has been off so long your profile picture got a promotion.β
βYour status update was in progress, then blocked, then somehow both at once.β
βPer my last email is the most polite way anyone has ever said please read.β
βYou have taken the last of the coffee so often the pot has your name on a wanted poster.β
βYou treat deadlines like weather forecasts: interesting, but not binding.β
βYour spreadsheet has more tabs than my browser and less structure.β
βYou have started three sentences on mute this week. That is a personal best.β
Use PG mode for the safest output. Keep jokes focused on tasks, habits, and office situations, and they will hold up in any channel.
You can describe a harmless behavior, but avoid private traits, protected classes, or sensitive personal details. Jokes about real named people are blocked.
The office-safe style is the most workplace-native, but classy and savage can make clean lines more surprising if you edit them before sending.
Roast the role, not the person: the 7 a.m. invite, the four-word Slack reply, the deck nobody read. That is the kind of line a boss can laugh at too.
Yes. Run a few PG-mode battles against an AI opponent. The judge scores wordplay and relevance, and the tip after a lost round shows what to sharpen.
Generate a line, battle an opponent, and turn the result into a replay your friends can judge.
Practice in PG Mode