← The Index  /  IDX_027

“I Tried Making $800 in 4 Hours with AI Agents (To See If It Works)”

Mikey No Code View the source ↗ Scored 2026-07-27 Engine 3.1 · prices 2026-07-20
73/100
Strained

4 of 6 axes hold. The mechanism exists; the claim strains somewhere the arithmetic can show you.

Six axes · 84 of 115 points

Temporal Possibility

Can the output fit the claimed window, and does the platform pay out in time?

20/20

The working

no unit count claimed; production-time check skipped

Workload Cost

What the claimed output actually costs to produce at current prices.

12/25

The working

unit count not stated; workload cost unverifiable — scored at midpoint

Stack Completeness

The full bill — named tools plus requirements the claim didn't mention.

9/15

The working

Base44: $30.0/mo (default)
ai_agent: $30.0/mo (default)
stripe_account: $30.0/mo (default)
stack total $90/mo across 1 named + 2 unstated

Price Realism

Claimed per-unit price against the real market benchmark band.

10/20

The working

neither unit price nor unit count stated; price realism unverifiable — scored at midpoint

Net Margin

Whether profit survives generation, tools, and time at market wage.

20/20

The working

no unit count; time cost $0
net: $800 - (gen $0 + stack $12 + time $0) = $788 (98%)

Replicability

Could a normal person with ~$500 replicate this, or does it need rare skill, an audience, or heavy capital?

12/15

The working

2 undisclosed requirement(s) raise the barrier (-3.0)
TIMEWORKLOADSTACKPRICEMARGINREPLIC
Six axes plotted against their maximum. Every point is the published score for that axis — the picture and the rows are the same numbers.

Extracted by qwen2.5:14b · scoring 3.1 · prices 2026-07-20. Scoring is deterministic: the same inputs reproduce this score, and every price is dated and sourced in the methodology. A hard fail overrides the total.

The Friday brief — free

Each week’s claims, scored, with the math.

Friday signal only. No guru fluff. — next drop Friday 8am PT