Build the small tool you always wanted
A working personal tool (tracker, calculator, mini-site) built by describing it, in an afternoon.
The problem
You have a spreadsheet-shaped chore or a tiny app idea, and "learn to code" was never going to happen.
What you get
A working system with the steps, tools, checkpoints, and expected return made explicit.
- Setup
- 180 minutes
- Back each week
- 2 hours
- Difficulty
- assisted
Expected return
The working case
A planning estimate, not a guaranteed result. Measure the first four weeks against your own baseline.
96
hours returned per year
At 2 hours/week across 48 working weeks.
2
weeks to earn back setup
Compare the setup estimate with the weekly time returned.
Operating contract
Input
The source material, constraints, and examples a human would need to do this work well.
Checkpoint
A person reviews judgment calls, sensitive content, unfamiliar tools, and irreversible actions.
Success signal
Track time returned, corrections required, and exceptions. Keep it only if the measured result compounds.
Before you start
- ·An afternoon and a concrete chore to kill
The steps
- 01
Describe the tool in one paragraph: what goes in, what comes out, who uses it. Feed it to an AI app builder.
- 02
Iterate in plain language: "make the button bigger", "add a weekly total", "it breaks when I enter zero". Screenshot errors back to it.
- 03
Stop at good-enough-for-you; personal tools do not need accounts, payments, or polish.
What it runs on
- Claude (artifacts)
Builds working mini-apps in the chat; free tier works.
- v0
For something you want deployed at a URL.
- Replit
When it outgrows a single page; AI agent included.
Where this goes wrong
- Do not put other people’s personal data in a vibe-coded tool; keep it personal-use until someone security-minded looks at it.
- Scope creep kills these; ship the one-feature version before adding the second feature.
Definition of done
Run it for four weeks. Then make it earn its place.
- □ Baseline the manual time before launch.
- □ Keep a human approval step for consequential output.
- □ Record corrections and exceptions, not just successes.
- □ Expand, revise, or retire it after the first review.
Build the system around it
Related workflows
If this one stops working, tell us. Three reports in a month and it leaves the library until a person has looked at it again.