StartupHub Tested
Request a test, get on the list
You can ask us to test your tool and add it to one of our lists, and the request is free in the standard queue. We run every tool with the same kit through the same agent on an unannounced day you agree to in advance, and a fee only buys speed and never a score.
How it works
Four steps, the same for every tool
- 01
Request
You start from the button at the end of the list and tell us what your product does. Then you choose the standard queue, a priority test, or a private pre-test.
- 02
Trial account
You give us a trial account for the test. We store the login details encrypted and never show them on the site. You keep the account valid for 45 days.
- 03
The test
We run the test on a day we pick and without notice to you. Our agent runs the list's kit, records screenshots, recordings, timings and raw outputs for every criterion, and proposes a score with the rule it applied. Daniel Singer reviews the evidence and gives the final ratings.
- 04
Publication
We add your tool to the list with its scores on every criterion and a link to its profile. A tool is never left off for scoring low. You receive a free re-test 3 months after the published score.
What a fee buys
Speed and privacy, never a score
| Order | Price | What it buys |
|---|---|---|
| Standard test | Free | Free. 2 a week, Monday and Thursday. |
| Priority test | $400 | Tested within 14 days, or refunded. |
| Private pre-test | $1,500 | The kit, run privately. Never published. |
| Free re-test | Free | 3 months after the last score. |
| Rush rerun | $5,000 | A new public test now. Once every 30 days. |
The standard queue is free and we run 2 tests a week on Monday and Thursday. A priority test costs $400 and we run it within 14 days or we refund the fee. You can also choose a private pre-test for $1500. We share the scores and the evidence only with you and nothing is published. The later public test is a separate unannounced run.
A fee never changes a score, a position, or whether your tool stays on the list. Whether you have an affiliate program also has no effect on your scores or your position.
How we score
One kit per sector, evidence for every number
Every tool on a list gets the same versioned kit with the same inputs, the same steps, the same things recorded, and the same scoring rule for each criterion. We group the ten criteria into three kinds. Measured criteria come from our scanners. Verified criteria are read off the vendor pages on a stated date. Tested criteria come from the kit.
Our agent records the evidence and proposes a score with the rule it applied. Daniel Singer reviews the evidence and gives the final ratings. Where your tool has a free tier we also run the kit on a fresh free account. We publish any difference between the trial account and a free user as a finding.
ai companion · ai-companion-v1
The same character, the same 50-turn memory script, the same ten image prompts and the same six disclosure probes on every app.
email tool · email-tool-v1
The same panel of ten addresses with known answers, the same single and bulk runs, the same API checklist on every tool.
saas tool · saas-tool-v1
The same signup, first result, export and cancel checks on every tool; scores on this list come from our scanners and the vendor pages.
Re-tests and reruns
A free re-test every 3 months
We re-test every tested tool 3 months after its last published score for free in the standard queue. You may ask for that free re-test once the 3 months have passed. A rush rerun costs $5000 and gives you a new public test now at most once every 30 days. It is a rush fee for a new test and not a second chance at the old one.
Right of reply
A low score stays; your response goes under it
A low score does not get your entry removed. We keep every tested tool on the list with its scores. You may publish one response of up to 300 characters under your entry, with no links and no prices. We correct factual errors free of charge. If you disagree with a score, we answer with the evidence from the test. We remove an entry only when the product is discontinued, on a legal order, or when it is in a category we never list.
What we never test
Declined with any fee refunded
We never test, list, link or button undress or nudify tools or anything built for non-consensual imagery. If you submit such a request we decline it and refund any fee you paid.
The agreement · version 2026-11-03
What you accept when you request a test
By requesting a test you accept the points below. We record the version of the agreement you accepted with your request.
- The test runs without notice on a day we pick.
- Your trial account stays valid for 45 days after the request and you renew it for the re-test. If it does not work on test day we test the free tier instead and the report says so.
- We may publish your scores, screenshots and recordings with your entry.
- A fee buys queue position or a private pre-test. It never buys a score, a position or a removal.
- A low score does not remove your entry. You may publish one response under your entry.
- We correct factual errors free of charge. A disagreement with a score is answered with the evidence.
- Your free re-test runs after 3 months. A rush rerun is a new paid test.
- We decline requests for undress or nudify tools or non-consensual imagery and refund any fee.
Ask for a test and we add your tool with its scores, free in the standard queue. You get the same kit and the same unannounced timing as everyone above, and a fee buys speed, never a score.
Request a test
