You know the feeling. Two options, both fine, and you have spent a week going back and forth instead of just picking one. It is not that the decision is hard. It is that you never gave yourself a way to decide.
I wrote a free skill for my AI agent that fixes this in ten minutes. One decision, two options, three scores, and you walk away with a choice and a first step. Not a big analysis. A finish line.
The mistake everyone makes
Most people try to think their way out of a stuck decision. They list pros and cons, ask three friends, sleep on it, then start the whole loop again the next morning. The problem is not missing information. The problem is no deadline and no scoring system, so every factor feels equally important forever.
The skill forces both. You write the decision as one sentence with exactly two options, and you give yourself a deadline. "Should I do A, or should I do B? I decide by Friday at noon." A decision with no deadline just sits there.
If you have three or more options, cross out the weakest one first. If you only have one option, your second option is "do nothing for now." That counts, and it is often the right answer.
Score it three ways
For each option, give it a score from 1 to 5 on three things only:
Money and time cost. 5 means cheap and quick, 1 means expensive and slow.
How much it helps. 5 means it clearly moves the goal forward, 1 means it barely matters.
How easy it is to undo. 5 means you can back out tomorrow, 1 means you are stuck with it.
Add the three scores. Here is a real example from the skill, a realtor choosing between two open house weekends:
Weekend A: cost 4, help 4, undo 5. Total 13.
Weekend B: cost 2, help 5, undo 3. Total 10.
Weekend A wins, even though B might bring slightly more foot traffic, because A is cheaper, faster, and easy to change. That is the whole point of the scores. They stop one flashy factor from running the decision.
But do not stop at the totals. Scores lie when one factor matters way more than the others. If one option wins on cost and help but is impossible to undo, say that out loud before you pick it.
The gut check and the tiny test
Two questions finish the job:
"If I had to pick in the next 30 seconds, which one would I pick?" That first answer is data, not magic. If it matches the scores, you are done. If it fights the scores, one of your scores is wrong. Find which one.
"What is the smallest test I can run this week?" Pick the version of the choice you can try for one day, one customer, or one post, then judge it on real results instead of guessing.
Then write it down:
Decision: one line.
Pick: A or B.
Why in one sentence: the reason that actually tipped it.
First step: the smallest thing you do today.
Test by: a date, one week out at most.
A decision you can explain in one sentence, with a first step you can do today, beats a perfect decision you never make.
What it does not do
This is the lite version on purpose. It does not compare five options at once, it does not dig into long term tradeoffs, and it does not handle high stakes calls like hiring, signing a lease, or spending money you cannot afford to lose. Those deserve a slower look and a second opinion from a real person. This is for the everyday stuck choices that are eating your week.
Links
- Free decision-helper-lite skill (GitHub): https://github.com/lilnickyai/claude-code-skills-samples/blob/main/decision-helper-lite/SKILL.md
- Free skills, one per week (email list): https://lil-nicky-ai.kit.com/d26c1ccfd6
- Universal AI Skills ($19): https://littlenova5.gumroad.com/l/leahgg
The free skill is complete on its own. If you want the full collection, with the deeper decision helper and everyday skills for ChatGPT, Claude, Grok, and 30-plus other AI tools, that is the Universal AI Skills pack.
Top comments (0)