Polish an email
Work & EmailWriting
State the recipient, your goal and the language, and get an email body written directly, with no reflexive apologies.
Fill the blanks and it copies as you filled it
[Paste the email you wrote, or what you want to say]
Turn the above into an email.
- Language:
- Recipient:
- What I'm after:
- Say things directly, and only apologize if I actually did something wrong
- One email body, nothing else0 of 3 blanks filled · anything left blank copies with its brackets
Describe the writing choice you want rather than a vague label like "polite". Setting a length means cutting padding, not writing stiffly. For writing that needs to be roundabout — a personal statement, an email to a professor — add a line about tone. The paper also reports that specifying a length shorter than the task needs backfires, because it drives follow-up questions (a 34% increase in 11% of scenarios).
What this prompt does
Compared with asking the same thing without the length and format lines.estimate
Estimated output tokens
700
Asked loosely
180
This prompt
Energy (Wh)
1.215
Asked loosely
0.423
This prompt
Token counts are estimated from character counts, not measured. Energy uses the per-band figures measured by Jegham et al. 2025 as published.
What that saves
15.8 Wh · 5.54 gCO₂e · 74 mL — assuming 20 uses a day
Tap a card for how the conversion was done.
Reported backfire — Specifying output lengthSpecifying a length longer than needed backfires (a 34% increase in 11% of scenarios)
Where these figures come from
We read the prompt to see whether it carries that instruction. The percentages below are published figures for the rule itself, not a measurement of what this prompt saves.
Assigning a role
44%mean reduction · 95% CI 29–59%
Podder et al. 2026 BP#4 · 19 use cases / 61 prompts
by modelGPT-4o 34%Command R+ 50%Mistral-7B 49%
Specifying output length
38.7%mean reduction · 95% CI 29.7–47.7%
Podder et al. 2026 BP#5 · 18 use cases / 57 prompts
by modelGPT-4o 35%Command R+ 49%Mistral-7B 32%
Reduced in every scenario. No backfire reported