// we set out to prove the agents were fake
We were wrong.
25 of 30 pass the Agent Test. We corrected ourselves 32 times getting there — 30 of them against our own thesis. The agents are real.
In 42 Claude Code sessions from one operator, the median uninterrupted run was 15 tool calls. The leash is you.
The leash is you
one real session, played backEvery leash number on this site comes from one sample: 42 Claude Code sessions · one operator · median 15 · max 92. It is a scale to read your own number against — not a benchmark, not a ranking, and not a claim about any other product or any other person. How it is counted →
4–5 · AGENT 2–3 · WRAPPER WITH AMBITION 0–1 · CRON JOB WITH EXTRA STEPS
Everybody has the tape
30 products · hover any cellThe receipts
| Product | Proof | Capability | Our verdict | Leash | The people say | Your call |
|---|
No product matches that combination. Nothing is hidden by default — clear the filters to see all 30.
Verdicts are our opinion of the shipped product vs its marketing, scored on the test above. Disagree? That's what the buttons are for. Wrong facts? Tell us.
Where we are weakest
our evidence, not the crowd's opinionThis is the list we would attack first if we wanted to prove this site wrong, so it is the list we publish. Community rankings — most disputed, biggest disagreement — appear here when there are enough votes to mean anything, and not before.
The Wrapper List
crowd-ruled, highest "not an agent" shareEvery verdict change, correction, and new judgment lands here with a date. Overturned calls stay on the record — that is the point of keeping one.