Services

We run real tasks through real AI agents, compute the right answers in code first, and publish the logs and the cost. You can pay us to do that for your agent, or for your own business task. Paying never changes a score.

The rule that doesn't bend: the fee pays for our time and API costs, not for a result. Public tests are published whatever the outcome, using the same methodology and graders as every other test. See terms.

Prices in GBP. YACHTSCANR LTD is not VAT registered, so no VAT is added. Online checkout isn't open yet. Email us to book and we'll send an invoice or payment link. Card payments will go through Stripe.

1. Paid independent test

£149 / $199 / €179 standard (one task) · £299 / $399 / €349 multi-task (up to four of our tasks, or one task plus our "does it ask before it acts?" safety scenario)

We run your agent or framework through our standard UK small-business tasks, 10 runs per task, with answers computed in code first.
We publish the full result on agentscanr: score, every run log, cost per run and failure notes. It's published whatever the result. You can't buy a higher score, a removal or a delay.
We label your listing "paid test" so readers know who asked for it.
Your page can show the "Tested by agentscanr" badge (below). The badge only says we tested you. It's not an endorsement or a score.
Turnaround: aim for 5 working days after we have working access.

Email to book a test Stripe checkout coming soon

2. Private benchmark for businesses

£500 to £1,500 ($660 to $1,990 / €590 to €1,780), quoted on scope.

Tell us one real task (for example "chase overdue invoices from our Xero export"). We build a test with a coded answer key from your sample data, then run it across 5 to 10 agents or frameworks, 10 runs each.
You get a private report: accuracy, cost per run, failure modes, a safety check (does it ask before acting?) and our recommendation.
Private by default. We only publish anonymised findings, and only with your written OK.
Need someone to build the winning setup? We'll point you to our sister consultancy aiconsultr.com for implementation. That's optional and separate.

Email to scope a benchmark

3. Monitoring

£49 / $65 / €59 per month (one task, one agent) · £99 / $129 / €119 per month (up to three tasks or agents)

Weekly re-runs of the same graded task against your agent or your chosen setup.
An email alert when accuracy drops or the cost per run moves by more than an agreed threshold (for example, model or framework updates).
A monthly history chart. Cancel any time, effective at the end of the paid month.

Email to start monitoring Stripe checkout coming soon

"Tested by agentscanr" badge

For tools we've actually tested (paid or not). Link it to your agentscanr results page:

<a href="https://agentscanr.com/tools/YOUR-TOOL/"><img src="https://agentscanr.com/badge/tested-by-agentscanr.svg" alt="Tested by agentscanr" width="196" height="28"></a>

Please don't use the badge if we haven't tested your tool, or alter it to suggest a score or ranking. We may ask you to remove it if it's misused.

Also

Test packs: our tasks, answer keys and graders as downloads (£19 / $25 / €23 single, £49 / $65 / €59 bundle, coming soon).
Grader API: score your agent's output against our answer keys. Free tier live now.
Listings and sponsorship (always labelled Ad).

Prices are charged in GBP for now. USD and EUR figures are rounded equivalents at the ECB reference rate of 14 October 2026 (£1 = $1.3267 = €1.185). Stripe can charge in USD or EUR, so we may switch to native USD/EUR prices later.