Pricing / AIpento

One app. Twelve tests a year.

Attackers are already pointing AI at apps like yours, and they do not stop after your audit week. Pro is $249 a month, billed yearly, for up to 10 apps: a full AI penetration test when you start, then a retest every month, so your app is tested twelve times a year instead of once. Need it done once? The one-time engagement is $2,499: an AI penetration test with AutoFix and Fix all for one app. Prices in US dollars.

Most popular

Pro

$249/ month

Billed yearly: $2,988 a year, for up to 10 apps.

For up to 10 apps you ship often and want watched all year, not once.

  • Up to 10 apps, tested 12 times a year
  • A full AI penetration test when you start, then a retest every month
  • Unlimited code scans between tests: SAST, secrets, vulnerable packages (OSV) and infrastructure files
  • Live tests mapped to the OWASP Top 10, API Top 10 and LLM Top 10
  • Every month you get the drift: what is new, what came back, what is still open
  • AutoFix and Fix all included, up to 25 fixes a month: each patch verified thrice from a clean checkout
  • Pentest report and attestation letter, refreshed every month
Get Pro

One-time engagement

$2,499one time

Paid once, for one app. Nothing renews.

For one app that needs a full test and every finding fixed, once: before a launch, a funding round or a customer's security review.

  • 1 app
  • A full AI penetration test, plus a retest to confirm the fixes
  • AutoFix included: every patch verified thrice from a clean checkout
  • Fix all: queue a fix for every open finding in one step
  • Code scans: SAST, secrets, vulnerable packages (OSV) and infrastructure files
  • Pentest report and attestation letter
Buy the one-time engagement

Enterprise

Quoted

Scoped and priced per customer, in writing.

For many apps, a one-off pass before an audit, strict access rules, or code that cannot leave your network.

  • As many apps as you need, tested on the cadence you choose
  • Single sign-on (OIDC), enforced two-factor, four roles
  • A tamper-evident, signed audit log you can export
  • Self-hosted runners, or the whole platform on your premises
  • An Evidence Pack mapped to SOC 2, ISO 27001, PCI DSS and OWASP ASVS, built from what was actually tested - a control no check covers is recorded as not tested, never as a pass
  • Priority support
Get a quote

What every plan checks.

In the words your security team and your auditor already use.

Access control

Reading or changing another account's data, acting above your role, weak sessions and tokens.

OWASP A01, A07 / API1, API3, API5

Injection

SQL, NoSQL, command, template and XML injection, and cross-site scripting.

OWASP A03

Leaked secrets

More than 200 kinds of API key, token and password written into code or configuration.

Secrets detection

Vulnerable packages

npm and PyPI dependencies checked against the public OSV vulnerability database.

SCA / OWASP A06

Infrastructure as code

Dockerfiles, Compose files, GitHub Actions workflows, Terraform and Kubernetes manifests.

IaC scanning

Misconfiguration and exposure

Missing security headers, exposed files and backups, open admin panels, debug endpoints.

OWASP A05

Server-side request forgery

Getting your server to fetch an address an attacker chooses, including internal ones.

OWASP A10

API abuse

Mass assignment, missing rate limits, unsafe use of third-party APIs, and the rest of the API list.

OWASP API Security Top 10

AI features in your app

Prompt injection, data leakage, poisoning and abuse of the language-model features you ship.

OWASP Top 10 for LLM applications

AI agent tools and MCP servers

The code that gives AI agents their tools, checked for actions they should never be able to take.

SAST

Your code, five languages

JavaScript, TypeScript, Python, Go and PHP, read line by line without being run.

SAST

Reliability and release

Swallowed errors, calls with no timeout, queries with no limit, missing lockfiles and loosely pinned versions.

Code quality

Every finding and its evidence is mapped to SOC 2, ISO 27001, PCI DSS and OWASP ASVS controls, ready for an auditor. That is mapped evidence, not a certification.

How the price compares.

Quoted per testa traditional penetration test is scoped and priced one assessment at a time, and the next one is quoted again
$249 a monthAIpento Pro, billed yearly: up to 10 apps, an AI penetration test plus a monthly retest - twelve tests a year. Or $2,499 once for a single engagement with AutoFix and Fix all.
Weeks to booka traditional pentest, and the report describes the app as it was on those days

Questions buyers ask first.

What counts as an app?

One web application and its API, tested on a staging address you own, plus the repository it is built from. Pro and the one-time engagement each cover one app; Enterprise covers as many as you need. If you are not sure, ask and we will tell you before you pay anything.

What is an AI penetration test?

AI agents do the work a penetration tester does: they map your app, plan the tests, run them against a staging copy you have authorized in writing, and link small weaknesses into real attack paths. A finding that cannot be reproduced is withdrawn. Every finding you see comes with the exact request and response that proved it.

What does verified thrice mean?

A fresh checkout, a fresh database, your full test suite. Then the same again from nothing, and a third time. Suites that pass once often lean on leftover data or test order, and a suite that passes twice can still be flaky. Three clean runs are what separate a fix from a lucky run.

Is Pro billed every month?

No. Pro is $249 a month, billed yearly: one payment of $2,988 for the year, for one app. It covers a full AI penetration test when you start and a retest every month after it - twelve tests across the year.

What happens in the monthly retest?

The same agents run again against your staging app and your current code. You get what is new since last month, what you fixed that has come back, and what is still open - plus a refreshed report and attestation letter. That is the point of the year: an app you shipped to in March is not the app we tested in January.

What does the one-time engagement include?

One app, one payment of $2,499, and nothing that renews. A full AI penetration test, a retest to confirm what was fixed, and AutoFix on every finding - each patch verified thrice from a clean checkout - with Fix all, which queues a fix for every open finding in one step. You keep the report and the attestation letter.

What access do you need?

Read-only access to the repository for code scans. For a penetration test, a staging address you own and a written authorization naming it, proved by a DNS record. We never ask for production credentials, and nothing runs against a system you did not name.

Which languages and stacks do you cover?

Code scanning covers JavaScript, TypeScript, Python, Go and PHP, npm and PyPI packages, Dockerfiles, CI workflows, Terraform and Kubernetes. Live tests work on any web app or API you can reach over HTTPS.

Can it run inside our network?

Yes. Enterprise includes self-hosted runners and an on-premises edition, with results in the same console as your cloud projects.

Which currency, and what about tax?

Prices are in US dollars. Any tax we are required to add is shown separately on the invoice.

Enterprise

Tell us about the app.

We reply with next steps and a fixed price, in writing, before any work starts. No call needed.

  • Scope and price written down before kickoff
  • Every finding proved, or withdrawn
  • An attestation letter you can hand to your auditor