You forget half the things you promised people today. Because the promises you forget are the ones that quietly cost you trust, and this catches them before anyone notices.
| Daily time cost | Under 10 minutes once the prompt is saved |
|---|---|
| Core prompt | List every commitment, who it’s for, and any deadline |
| Habit trigger | Same time daily, e.g. right before closing your laptop |
| Trust risk | Failing to follow through on commitments erodes workplace trust (SHRM) |
1. Gather today’s outgoing messages
At the end of your work day, pull together everything you sent: emails, Slack DMs, WhatsApp texts, whatever channel you use most. You don’t need every message, just the ones where you talked to other people about work. Copy the text as is, typos and all.
2. Give AI one exact instruction
Paste the messages into your AI assistant with this prompt: ‘Read these messages I sent today. List every commitment I made to someone else, who it is for, and any deadline I mentioned. Flag anything with no deadline as needs a date.’ Save this prompt somewhere you can reuse it daily.
3. Get your checklist
AI reads through the noise and pulls out just the promises: what you said you’d do, for whom, and when. It skips the small talk and the context you don’t need. You get a short list, not a summary of your whole day.
4. Cross off, carry forward
Check what’s already done. Whatever is left, either do it now or carry it into tomorrow’s list. This takes two minutes and it is the step that actually saves you from broken promises.
5. Make it a five o’clock habit
Set a daily reminder for the same time each day, right before you close your laptop. The whole thing takes under ten minutes once the prompt is saved. Do it for a week and you will stop wondering what you forgot to send.
A worked example
Priya manages three client accounts and makes small promises all day over email and Slack without writing any of them down.
Before
Sure, I’ll send the updated deck by Wednesday.
Let me check with finance and get back to you.
I’ll loop Rahul in on this thread.
Happy to hop on a call next week, I’ll send you some time options.
After
1. Send updated deck to client, due Wednesday
2. Check with finance on budget approval, no deadline given, follow up today
3. Loop Rahul into the client thread, not done yet
4. Send call time options to client, due this week
Three ways to bend it
For managers tracking a team
Instead of your own sent messages, paste the promises your team made in a shared Slack channel or client thread. Ask AI to group them by team member and deadline. Now you have one glance at what the whole team owes people this week.
For salespeople and follow ups
Run the same prompt on emails you sent to prospects. Ask AI to separate promised to send something from promised to follow up by a date. This turns a messy inbox of sales chatter into a simple follow up list.
For freelancers juggling clients
If you work across multiple clients, ask AI to group the checklist by client name instead of by date. That way you see exactly what you owe each person, which is what actually gets asked about on your next call.
When it goes wrong
AI misses a promise I made casually, like ‘I’ll get to it soon’
Add a line to your prompt asking it to catch vague commitments too, not just ones with clear deadlines. Give it one or two examples of how you usually phrase things so it learns your style.
The checklist is too long and just feels like more work
Narrow the input. Only paste messages to people outside your immediate team, or only from your top few clients. The goal is catching what you’d actually be embarrassed to forget, not logging everything.
When not to use this
If you already write down every commitment the moment you make it, in a task manager or notebook, skip this. This technique is for people who promise things in the flow of a conversation and move on before it’s written anywhere, which is most people on a busy day.
What you can now do
You stop finding out three days later that you forgot to send something you promised.
Part of Making Life Easier With AI, a weekly walk-through of one small thing AI can take off your plate.
Take this further
Act as a blunt hiring manager who has read ten thousand resumes, not a career coach. I will paste my full resume and the job description I want. Rewrite the whole resume for that role, section by section, in this order: summary, experience, skills, education. Rules: every experience line leads with impact, not duty. Use bracketed placeholders like [8 percent] for any number I did not give you, and list at the end every placeholder I need to replace with a real figure. Keep it to one page of text. Plain formatting only, no tables or columns, so screening software can parse it. After the rewrite, tell me the three weakest claims that need evidence before I send this anywhere. My resume: [paste resume]. The role: [paste job description].
Takeaways
- Save the exact extraction prompt in a notes app so you paste it in unchanged every day instead of rewriting it.
- Set a recurring reminder for the same time each day, right before you close your laptop, so the check becomes automatic.
- If the checklist feels too long, narrow what you paste in to messages sent outside your immediate team or to your top few clients.
- Add a line to your prompt asking AI to flag vague, undated commitments like ‘I’ll get to it soon,’ not just ones with explicit deadlines.
- If you manage a team, paste the shared channel history instead of your own messages and ask AI to group results by person and deadline.
Questions people actually ask
How do I get AI to pull the promises out of my emails and Slack messages?
Copy the messages you sent that day, typos and all, into your AI assistant. Ask it to list every commitment you made, who it was for, and any deadline mentioned, and to flag anything without a date as needing one. Save that exact wording so you can reuse it every day without rewriting it.
What if AI misses a vague promise like ‘I’ll get to it soon’?
Add a line to your prompt telling the AI to catch casual or vague commitments, not just ones with explicit deadlines. Giving it one or two examples of how you typically phrase soft promises helps it learn your style and catch similar phrasing next time.
Is this technique only useful for people who manage a team?
No. It started as a personal end of day check, but it adapts to managers, salespeople, and freelancers by changing what gets pasted in. Managers paste team channel messages, salespeople paste prospect emails, and freelancers ask AI to group results by client instead of by date.
How long does this actually take once it’s set up?
The full routine, gathering messages, running the prompt, and crossing off what’s done, takes under ten minutes once the prompt is saved somewhere reusable. Most of that time is spent deciding what to do with each item, not waiting on the AI to produce the list.
Do I still need this if I already write commitments down in a task manager?
No. This is meant for commitments made in the flow of a conversation and never written down anywhere, which is what causes them to be forgotten. If you already log every promise the moment you make it, you have already solved the problem this technique addresses.
Sources
- Prompting best practicesAnthropic
- Why Trust Matters at WorkSHRM
- Zeigarnik effectWikipedia
What happens next
As more people build daily habits around pasting message logs into AI assistants, expect email and chat clients to start offering commitment extraction as a built in feature instead of a manual copy and paste step. The weak point to watch is accuracy on vague, undated promises, since that is where users still report AI missing things. Whether people trust the AI generated list enough to stop mentally double checking it will determine if this stays a niche habit or becomes a default workflow.
