Start with a free preview. Then choose how deep to go.

Every package starts with a free preview of how AI agents do on your product. We quote the full audit after we go through the preview with you.

Get your free preview report

From request to findings.

  1. Day 1

    Request your free preview

    Tell us about your product and your customers.

  2. Week 1 to 2

    We put an agent to work

    An AI agent uses your product the way your customers' agents would.

  3. Within 2 weeks

    Your preview arrives

    One major finding in full. Your other findings and fixes, blurred.

  4. After the preview

    Choose a package

    We go through the preview with you and quote the package that fits.

Pick the depth you need.

Included SilverThe audit GoldThe audit and its evidence PlatinumAll three, against your category
Agent Audit
Two review meetings
Evidence Pack
Competitor Report
BuySilverBuyGoldBuyPlatinum
PriceQuoted after your free preview

What each package holds.

SilverGoldPlatinum

Agent Audit

What happened when an AI agent used your product, and what to fix first.

Buy Silver
  • What the agent didEvery page it read, request it sent, response it got and retry it made, in order.
  • Where it failedThe exact point the agent stalled, retried or quit.
  • Why it failedThe cause of each failure, with the request and response that show it.
  • What it costCredits, tokens and time spent on each failure, so you can fix the expensive ones first.
  • What workedCapabilities the agent used well that your docs don't take credit for.
  • Two review meetingsIn the first, we go through the report with you and note your concerns. We work on them, then meet again to go through what we found.
GoldPlatinum

Evidence Pack

The audit tells you what happened. The Evidence Pack gives you the files it was built from, for every feature we tested.

Buy Gold
  • Integration codeThe working integration the agent wrote for each feature.
  • Session transcriptsThe full log of every session.
  • Raw API responsesThe unmodified request and response behind every finding.
  • Agent self-reportsWhat the agent said about its own work.
  • Failed sessionsEvery session that didn't produce a working integration, and why.
  • Session metadataCost, timing and outcome for each session, in a format your team can load into its own tools.
Platinum

Competitor Report

The same tasks, under the same conditions, across the other vendors in your category. You see where you stand and why.

Buy Platinum
  • Where you rankYour result beside each competitor's, per capability.
  • Where they winTasks a competitor finished that the agent failed on your product, and what their docs or API did differently.
  • Where you winTasks only your product got right. These are claims you can make with evidence behind them.
  • How they failedThe failure types that recur across your category, so you can avoid them before an agent finds them.

Independent by design.

We pay like any customer

We sign up and buy our own key, so we see exactly what your customers' agents see.

No one buys a better score

Vendors can't pay for a better result or a better rank, including our own customers.

See where agents fail on your API, free, within 2 weeks.

Get your free preview report

No commitment. We quote the full audit after the preview.

  • No access to your systems
  • Free, within 2 weeks
  • Findings shared only with you

Get your free preview report

Leave your work email and company name. We'll email you to arrange your free preview.

No commitment. Ready within 2 weeks.