Chat pleasantries in a system prompt
Sign-offs written for a human conversation, left in a machine instruction.
Phrases like "let me know if you need anything else" belong at the end of a message to a person. In a system prompt they are instructions to be chatty, which costs tokens twice: once to send them and again in the longer output they encourage.
The output cost is usually the bigger one. Output tokens are priced five to six times higher than input tokens on most models, so an instruction that adds two sentences of pleasantry to every response is expensive in a way that is easy to miss.
If you want a conversational tone, say so once in a tone instruction. Do not model it by example inside the instructions themselves.
Answer the question. I hope this helps! Let me know if you need anything else.Answer the question.Check your own prompt
The analyser checks this pattern along with the other 25, prices each finding against your request volume, and hands back a rewritten prompt. It runs in your browser — nothing is uploaded.
Run the analyserMore on filler
- A role that describes nothing"You are a helpful AI assistant" tells the model nothing it did not already assume.
- Prompt folkloreTips, deep breaths, and other emotional-appeal phrasing, copied from 2023-era prompting advice and never removed.
- Politeness aimed at a billing meterPlease, thank you, and "I would like you to" cost tokens on every request forever.
- Long ways of saying short things"Due to the fact that" is four tokens. "Because" is one.