Alternatives
Octomind, Bugster and DogQ shut down: where to move your tests
Three AI testing tools closed in 2026. If your tests ran on one of them, this page covers what happened, what you can still save, and your options, including the ones that do not involve us.
Facts checked on . More tools changed status this year: the AI QA Directory status page keeps a dated list.
At a glance
| Tool | What happened | What you can still keep |
|---|---|---|
| Octomind | Octomind turned its product off at the end of May 2026 and wound the company down by the end of June 2026. Today octomind.dev no longer resolves, and almost all of its public GitHub repositories are archived. | Playwright files, if you exported them while the service ran: Octomind's command line tool could write a test's Playwright code and config to a folder. That tool talked to Octomind's app, which is gone, so the route is closed now. The farewell letter also promised customers direct help with exports. |
| Bugster | Bugster's letter of May 22, 2026 says the company decided to close. Its sign-in page now says the service shut down on May 4, 2026, and its public GitHub repositories are archived. | Your test specs, if they are in your repository: the YAML files in .bugster/tests describe each flow, its steps and the expected result. Only Bugster's agent could run them, so they are a precise list of what to rebuild, not tests that still run. |
| DogQ | A notice on DogQ's website, as recorded in search engine copies and by the AI QA Directory, said it had stopped taking new accounts and would close fully on September 1, 2026. Today dogq.io and its documentation redirect to another company's website. | Little: the DogQ documentation we could check describes no export of tests, only a PDF export of run statistics. Screenshots, step lists and selectors you saved are still a good start for a rebuild. |
Octomind
What it was
An AI agent platform for end-to-end tests of web apps. Tests were created from a prompt, recorded in the browser or suggested by the agent from a URL. Octomind stored each test as a series of steps and generated standard Playwright code from them just before each run, in its cloud or locally. An AI auto fix feature, in beta, repaired failing tests.
What happened
Octomind turned its product off at the end of May 2026 and wound the company down by the end of June 2026. Today octomind.dev no longer resolves, and almost all of its public GitHub repositories are archived.
We've decided to close Octomind. The company will wind down by the end of June, and our product will be turned off at the end of May.
What to do this week
- Look for exported Playwright files: a folder written by the command line tool, or whatever Octomind's team sent you.
- Run them once locally against your test environment and note which tests fail.
- Put the tests that pass in your repository and run them in your CI on every pull request. With nothing exported, list your flows and rebuild the riskiest first.
Sources: Octomind's documentation, archived on GitHub, AI QA Directory fact sheet on Octomind.
Bugster
What it was
AI testing for web apps. Each test was a YAML file of plain language instructions (the page, the steps and the expected result) kept in a .bugster/tests folder in your repository, and Bugster's AI agent ran them in a browser. There was no Playwright code.
What happened
Bugster's letter of May 22, 2026 says the company decided to close. Its sign-in page now says the service shut down on May 4, 2026, and its public GitHub repositories are archived.
We've decided to close Bugster. The product is winding down, and the company will follow shortly after.
What to do this week
- Check your repository for a .bugster/tests folder, and keep it.
- Rank those flows by how much a failure would hurt, so the riskiest ones are rebuilt first.
- Rebuild them as Playwright tests in your own repository, so the next vendor change cannot take them away.
Sources: Bugster documentation on how tests are stored, Bugster's sign-in page, AI QA Directory fact sheet on Bugster.
DogQ
What it was
A web testing platform without code: tests were scenarios of steps built in DogQ's web editor with CSS or XPath locators, and AI generated steps from a prompt, suggested scenarios and diagnosed failing runs.
What happened
A notice on DogQ's website, as recorded in search engine copies and by the AI QA Directory, said it had stopped taking new accounts and would close fully on September 1, 2026. Today dogq.io and its documentation redirect to another company's website.
What to do this week
- Collect what you still have: screenshots, step lists, selectors and bug reports.
- List your user flows and rank them by risk.
- Rebuild the riskiest ones first, as code in your repository that anyone can run.
Sources: AI QA Directory fact sheet on DogQ, AI QA Directory status page.
Your options
Three ways forward
Whichever you choose, keep your tests as code in your own repository.
Run the tests yourself
Playwright is free and open source. Exported Playwright tests run in your own CI, and our guide to Playwright on GitHub Actions has a working setup. Playwright also comes with three test agents: a planner that explores your app and writes a test plan, a generator that turns the plan into tests, and a healer that runs the suite and repairs failing tests.
Switch to another AI testing tool
The AI QA Directory lists the agentic end-to-end testing tools, with dated facts and sources. Before you sign up, ask the questions below.
Hand the work to a managed service
That is what we do. Claude agents write and maintain Playwright tests in your repository, and a QA engineer reviews every change and signs a Go or No-Go call before each release. US$2,000 a month for up to 500 user flows, starting with a 60-day pilot at US$1,000 a month.
Questions to ask any testing vendor
A shutdown costs least when the answer to the first three is "in my repository", "yes" and "yes".
- Where do my tests live, and in what format?
- Can I export them as standard code today, not only when you close?
- Do they run in my own CI without your service?
- Who reviews a test after the AI changes it?
- What happens to my tests and my data if you shut down?
Moving to Releaseworthy
How we take over
- Send what you have. Exported tests, a list of flows, or just your product and its test environment. Within 2 working days you get a plain answer: what runs, what is broken and what we would cover first. This migration check is free.
- Start small, if you like. The Release Risk Audit takes 5 working days and costs US$490: we rank your flows by risk and rebuild or repair the 5 most critical ones as Playwright tests in your repository, running in your CI.
- Or start the full service. The 60-day pilot costs US$1,000 a month: every flow that matters, a person reviewing every failure and a Go or No-Go call before each release.
FAQ
Questions about moving
Can you take over the Playwright tests we exported from Octomind?
Yes. They go into your repository and run in your CI. We fix the ones that fail because the product or the setup changed, and a person reviews every fix. The free migration check tells you what that involves before you pay anything.
We used Bugster or DogQ and exported nothing. Where do we start?
With your user flows. Bugster kept its test specs as YAML files in your repository, so check for a .bugster/tests folder first: it lists your flows already. Otherwise the first step is to map your flows and rank them by risk, then rebuild the most critical ones as Playwright tests in your repository.
Do we have to use your service?
No. Running Playwright tests yourself is free, and the options on this page include tools that are not ours.
Are you affiliated with Octomind, Bugster or DogQ?
No. Releaseworthy is not affiliated with any of them. The facts on this page come from their own announcements where we could find one, and from the AI QA Directory, each one linked and checked on October 19, 2026. The product names belong to their owners.
Tell us what you are shipping next
One email is enough to start: your product, your platforms and how often you release. You get a plain answer on fit and, if it fits, a written proposal for the 60-day pilot with a short video of how we would test your product.
Or write to qa@releaseworthy.app. Still deciding after the proposal? We book a Google Meet call for your questions.