QA.tech

What it does

QA.tech takes a plain-English description of a user journey or API outcome and tests it against a live web or native mobile app, an API endpoint, or a pull request with a preview environment. It returns pass or fail results with screenshots, logs, network activity and the agent's reasoning at the point of failure. For APIs it makes real HTTP calls and returns the full request trace.

Teams use it to cover flows that previously lacked testers, run dynamic checks on every pull request, and read failure breakdowns in GitHub. Tests are created from chat rather than scripts, and the MCP server lets Cursor, Claude Code or Codex start runs and read results. The agents are vision-based and use the UI as a person would, without selectors.

It suits QA teams, engineers, CTOs and DevOps teams shipping web, mobile or API products. Starter is self-serve at $89 per month with 50 credits and up to 10 users, while Growth and Enterprise are custom-priced. Mobile and API testing are not included on Starter. API testing is functional and contract-style, not load or penetration testing, and no source-code access is required.

What it can do

• Plain-English test creation that generates steps automatically
• Dynamic PR testing against preview environments with GitHub comments and failure breakdowns
• Mobile testing for iOS and Android apps via emulators and simulators
• API testing with full request and response traces and sandboxed validation code
• MCP server exposing 16 tools to Cursor, Claude Code and Codex for running and creating tests
• Debugging evidence including screenshots, console, network activity and reasoning

Works with

• GitHub
• GitLab
• Bitbucket
• GitHub Actions
• CircleCI
• Jenkins
• Jira
• Linear
• Slack
• Microsoft Teams

Before you choose

• Mobile and API testing are not included on the Starter plan.
• Growth and Enterprise pricing is custom and requires talking to sales.
• API testing is functional and contract-style, not load or security testing.

Listed under

• Coding
• Code review and testing
• Ci integration
• Engineering teams
• Mobile testing
• Pull request testing
• QA testing
• Web testing

Get started

Demo is the stated entry route; Starter has a 'Try now' button.

1. Connect your application or API and a source such as GitHub or a preview environment.
2. Describe user journeys, edge cases and domain rules in chat so the agent learns them.
3. Connect Claude Code, Cursor or Codex to the MCP server to run tests from the editor.

Starter is $89 per month with 50 credits included; only test executions count toward plan usage. Growth and Enterprise require contacting the company.

Source: qa.tech/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Dynamic Testing on Every PR – AI QA Testing Tool for Web & Mobile ↗qa.tech
• AI API Testing – Test REST & HTTP APIs Without Scripts | QA.tech ↗qa.tech
• MCP Server for AI Coding Agents – Cursor, Claude Code & More | QA.tech ↗qa.tech
• AI QA testing pricing and plans | QA.tech ↗qa.tech
• AI Agents - QA.tech Docs ↗docs.qa.tech

Company website

qa.tech: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.