Free
Try it on one site.
$0/mo
- 3 test runs (total)
- 1 project
- Jev in every run
- Live browser view
- Findings in your editor (MCP)
- Community support
Pricing
Start free. Scale when you ship.
Every plan runs the same real browser driven by the same frontier models. The difference is how many runs you get and how many people share them.
The short answer
ZeusQA is free for three test runs with no card. Pro is $39 a month for 300 runs, unlimited projects, 10 schedules and optional session video. Team is $149 a month for 1,500 runs, 50 schedules, organizations, roles and shared projects. A run is one test against one URL, however many steps it takes.
Pro
Most popularFor one developer, testing daily.
$39/mo
- 300 test runs / month
- Unlimited projects
- Jev in every run
- Live view + session video
- Security review of pull requests (100 commits / month, 20 / day)
- Vulnerability watch on 10 repositories, with fix pull requests and Slack or Discord alerts
- 10 schedules
- Findings in your editor (MCP)
- Priority models (Astra)
- Email support
Team
For a team sharing one QA budget.
$149/mo
- 1,500 test runs / month
- Organizations with members
- Roles & shared projects
- 50 schedules
- Security review of pull requests (500 commits / month, 20 / day)
- Vulnerability watch on 50 repositories
- Everything in Pro, Jev included
- Priority support
Secure payments by Stripe. Cancel any time from the billing portal. Prices in USD.
What every plan includes
- A real Chromium browser, driven step by step by the model you pick.
- Any vision model on LLM Gateway that can drive a browser, chosen per run.
- Laptop, tablet and phone sizes with real touch and mobile user agents.
- A verdict with named checks, severity-rated findings and a screenshot per step.
- Session video on Pro and Team, off by default: turn it on for one run or for every run in Settings.
- Console errors, uncaught page errors and failed requests collected on every run.
- Passive security checks on every run: headers, cookies, mixed content, open redirects, source maps, leaked secrets and stack traces.
- Live browser view while the run is happening.
- An MCP server: Claude Code, Cursor or any MCP client reads your runs and findings, starts tests and checks its own fixes, with an organization API token.
What is not included, on any plan
Worth knowing before you compare us to something else.
- No CLI. A run starts from the app, a schedule on Pro or Team, or a pull request through the GitHub App.
- Schedules run at most once an hour, and alerts are email only — no Slack, SMS or webhooks.
- Chromium only — no Safari or Firefox.
- No visual regression or pixel diffing against a baseline.
- No public, shareable run links. Runs sit behind your account.
Questions
What counts as a run?
One test against one URL, however many steps it takes. A run that ends blocked still counts, because it used a real browser and a real model.
Do free runs reset?
No. Free is three runs in total, not three a month. It is there to let you see a real report before paying.
What happens when I run out?
Runs stop until the next billing period, or until you upgrade. Nothing is deleted and no run is charged for twice.
Can I change plan later?
Yes, from Billing in the app. Upgrades take effect immediately; downgrades take effect at the end of the period.
Is there a free trial of Pro?
No trial, but the three free runs use the same browser and the same models as a paid run, so they are representative.
Which plans include schedules?
Pro includes 10 schedules and Team 50, each running hourly, daily, on weekdays or weekly in your time zone. Scheduled runs count toward the plan like any other run. Free has no schedules.
How many commits does the security review read?
Pro reviews up to 100 commits a month and Team 500, at most 20 a day per workspace and 20 an hour per repository. Reviews are also shared across ZeusQA, up to 40 a day for all workspaces together. A commit past a limit gets a check that says which one; while the repository's merge gate is on, that check does not pass (it blocks a merge where the check is required), except past ZeusQA's shared 40 a day: that capacity is not your workspace's, so the check stays neutral. Re-running the check reviews it once the limit resets. A commit replaced by a newer push before its review starts, or a review that fails, does not count toward the month; every commit queued counts toward the day and the hour. The review is off for each repository until you turn it on in its settings, under Connections, GitHub.
Does the security review find secrets?
Yes. Before the model reads the diff, fixed patterns check every added line of each file GitHub returns a diff for, up to the first 300 files and lockfiles left out, for cloud and provider keys, private keys, access tokens, connection strings with a password and random-looking values assigned to secret names. Each match, up to 30 per commit with a count of the rest, is a finding on its line that shows only the key's prefix, rated low in test, mock and example files, and the value is masked before the model sees the diff. A secret goes through the repository's merge gate like any other finding: at the gate's severity or above it fails the check.
What does vulnerability watch do?
On Pro and Team, a connected repository can have its root manifests (package.json with its lockfile, requirements.txt, go.mod, Cargo.lock, composer.lock, Gemfile.lock) checked once a day against OSV.dev and CISA's Known Exploited Vulnerabilities catalog. Each advisory gets a briefing on the Advisories page, a direct dependency with a fixed version gets a pull request that bumps it (up to 3 a day per repository and 10 open at once), and new advisories go to the Slack and Discord webhooks you add in Settings, Notifications. Pro watches 10 repositories and Team 50. Nothing in the repository runs, and the pull requests need the GitHub App's Contents permission at Read and write.
Do you offer annual billing?
Not today. Both paid plans are monthly and cancel any time from the billing portal.