EasyEnv

How sharp are your prompts? Prove it on the leaderboard.

One spec, one model, one prompt to get right. The best scores land the exact output in the fewest tries. Get a score out of 100, a tier, and your rank.

Free to play. Runs in your browser. Sign in to save and share your rank.

Prompting ScoreExample
0
out of 100
Advanced
Top 6%
percentile
Matched
challenge

Exact output on the third try. Tight constraints, no rambling.

The challenge

Hit the exact spec

The model has to return strict JSON in a fixed shape, and right now it does not. This is the actual console you get. Shape the prompt until every field validates, in as few tries as you can.

Prompt Engineering Challenge
57:20 left
schema.jsonprompt.txt
1Extract the invoice.
2
3Return the total and the due date.← too vague
4
5# The model keeps adding prose and mislabels the date.
Terminal
$ run prompt.txt
model > Sure! The invoice total is $420 and it is due...
validate > expected JSON object, got prose
validate > field "due" not found
FAILED (0/2 fields)
One real model60 minute windowStrict output specFewer tries score higher

What we score

It rates your prompting, not the model

Everyone gets the same model and the same spec. The score is your craft: did the output match exactly, did you constrain it well, and how few attempts it took.

1

Match the spec

The output has to validate against the exact shape asked for. Close is a miss.

2

Constrain it

We reward prompts that pin down format, tone, and edge cases so the model cannot wander off.

3

Fewer tries

Landing it in two attempts beats brute-forcing it in twenty. Precision over trial and error.

The leaderboard

Where will you land?

Every run is judged on the same challenge, so the score is comparable. Climb from Learning to Elite, beat your friends, and share a link that shows exactly where you rank.

Claim your rank
90+
Expert
Fast, clean, every trap caught
90 - 100
75+
AdvancedExample run
Solid work, challenge shipped
75 - 89
55+
Proficient
Got there, a few detours
55 - 74
35+
Developing
Partly there, kept going
35 - 54
0+
Beginner
Made a start, room to grow
0 - 34

How it works

Three steps, one hour, one rank

01

Boot the challenge

A real Linux machine spins up in your browser with the model, a prompt file, and a validator. Nothing to install.

02

Shape the prompt

You have an hour. Edit the prompt, run it, read what comes back, and tighten it until the output matches the spec.

03

Get scored and ranked

When the output validates, the judge scores your run out of 100, hands you a tier, and drops you on the leaderboard.

Find out how sharp your prompts really are

One spec, one hour, one shareable rank. No setup, no install. Start in your browser right now.

FAQ

Questions, answered

It is a hands-on challenge on a real Linux machine. You shape one prompt until a real model returns the exact output the spec demands. You have up to an hour, and we score the outcome, whether you found the real root cause, and how clean your change was.

No. The whole challenge runs in your browser on a real Linux box with an editor and a terminal already set up. Nothing to download, nothing to configure.

You have up to an hour, but the highest scores finish fast and clean. It is free to play. Sign in to save your result and share it.

A shareable score out of 100, a tier, and your rank against everyone else who has taken it. Send the link to a friend and challenge them to beat it.

Prompt Engineering Challenge: Test how well you prompt | EasyEnv