Turn recurring writing feedback into a narrow editing coach: use authorized guidance and your own before-and-after examples, define what must stay true, and review a focused edit before sending the message yourself.
Amir turns repeated feedback that his messages are too long into a custom GPT, grounds it in specific writing guidance, adds constraints to preserve the user's voice, and uses it while drafting Slack messages.
A custom GPT that returns a concise revision, explains the important edits, and preserves the writer's facts, intent, terminology, and recognizable voice.
OpenAI conversational AI
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
4 steps
Translate the feedback into a concrete editing target such as message length, lead placement, action clarity, or readability. Write a small set of principles from material you may use and pair them with your own examples.
The quality of your AI coach depends on the quality of the expert material you provide it. Use trusted sources.
Configure the GPT as a focused editor. Tell it the audience, task, protected facts, and output format. Make clear that it should preserve the writer's voice and leave good text alone.
Edit workplace messages for [specific goal]. Preserve facts, names, numbers, dates, decisions, terminology, tone, and action owners. Make the smallest useful edit. Return the revision plus a short note on what changed. If the original already meets the goal, say so and keep it mostly intact.
Add explicit constraints based on the writer's habits: target length, preferred structure, words to avoid, acceptable bullets, and when context must remain. Include before-and-after examples instead of generic style adjectives.
Follow these constraints: [constraints]. Match these paired examples: [examples]. Do not add headings, bullets, enthusiasm, certainty, or calls to action that the original does not contain.
Test with synthetic messages before using real work. Compare facts and actions, read the revision aloud, make the final edit yourself, and update the coach when the same correction recurs.
Revise this message for [audience and channel]. Protected facts and actions: [list]. Target: [goal]. Text: [message]. Return the edited version and flag any ambiguity you could not resolve.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Log your workouts effortlessly without ever opening a spreadsheet. This workflow uses the Claude mobile app to parse a simple voice note and automatically update your workout tracking spreadsheet in the background.
Create a custom email management interface within Claude to read, triage, and draft replies without opening the Gmail tab. This workflow helps you reclaim focus and compound your learnings by training your AI on every interaction.
Build a fully automated system in Claude to generate personalized, interactive HTML proposals and client onboarding materials, saving hours on administrative work and creating a high-touch experience.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.