agent-qa

What is agent-qa?

agent-qa is an AI testing agent built by vostride for teams that want end-to-end checks without hand-writing browser scripts. You describe a test in plain English inside a YAML file, and the agent drives a web page through Playwright or a mobile app through Appium, then reports a verdict with recorded evidence. When a step fails, it looks at the screen again and tries another route to the same result. A memory layer called memidex stores what earlier runs proved about your app as reviewable files, and it can replay actions that already worked to save tokens and runtime. Everything lives in your repository, and coding agents can author or run tests through the CLI, an MCP server or installable skills. You bring your own model credentials and your own execution infrastructure. According to the vendor there is no paid tier, hosted lock-in or seat fee, and the source is available under a source-available licence rather than a classic open-source one.

Use cases

• Test web apps in plain English
• Run mobile app tests on Android and iOS
• Add AI regression checks to CI
• Let a coding agent verify its own changes
• Keep UI tests alive through layout changes
• Collect screenshots and evidence per run

Who it's for

• QA engineers
• Software developers
• Engineering teams

Why use agent-qa

• Tests are written as readable plain-English steps, so people who do not script browsers can contribute to them.
• Steps adapt when the interface changes, which cuts the maintenance that selector-based tests usually need.
• Tests, memory, cache and rules stay as files in your repository, so teammates and CI review and revert them like code.
• One flow can cover web, Android and iOS instead of keeping separate suites per platform.
• You pick the model provider and pay no vendor fee to the maker.

agent-qa features

• Plain-English tests written as YAML files
• Web testing through Playwright
• Android and iOS testing through Appium
• Self-healing steps that re-observe the screen after a failure
• memidex persistent memory stored as reviewable files
• Action caching to cut tokens and runtime
• Sandboxed Node, Bun, Python or Bash hooks during tests
• Bring-your-own LLM provider, endpoint or model in config
• CLI, MCP server and installable skills for coding agents
• Recorded test evidence for each run

agent-qa integrations

• Playwright
• Appium
• Claude Code
• Codex
• MCP

How to get started with agent-qa

1. Install the npm package with npm install -D agent-qa and run npx agent-qa init for your platform.
2. Configure a model and a target app, either by hand or by handing the setup prompt to your coding agent.
3. Write a test as a plain-English YAML file, or have a coding agent author it through the MCP server.
4. Run it locally or in CI; the agent executes each step, retries another route if one fails and records evidence.
5. Review the verdict and the lessons memidex proposes, then keep or revert them as ordinary file changes.

agent-qa limitations

• You supply your own model credentials and execution infrastructure
• Source-available licence rather than a standard open-source one, so read it before commercial use
• Published benchmark scores are self-reported and cover mobile task completion, not QA correctness on your app

agent-qa pricing

agent-qa is free to use.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

agent-qa alternatives

The main agent-qa alternatives listed on Debuly are Playwright, Stagehand, mabl, QA Wolf, Applitools and testRigor. The table compares pricing and platforms; order is A–Z.

1. 

   Applitools

   Visual AI testing platform for catching UI and functional bugs across browsers and devices, with a 14-day free trial.

   CodingDesignAutomation and agents
   Free trialVisit Applitools (official site)
2. 

   mabl

   Agentic test automation platform that builds, runs and self-heals end-to-end web, mobile, API, accessibility and performance tests.

   CodingAutomation and agentsDeveloper platforms
   Free trialVisit mabl (official site)
3. 

   Playwright

   Open-source Microsoft framework for end-to-end web testing and browser automation across Chromium, Firefox and WebKit, with CLI and MCP for agents.

   Developer platformsCodingAutomation and agents
   Open sourceVisit Playwright (official site)
4. 

   QA Wolf

   AI-driven end-to-end testing for web, mobile and desktop apps, with tests written in Playwright and Appium and billed per run.

   CodingAutomation and agentsDeveloper platforms
   PaidVisit QA Wolf (official site)
5. 

   Stagehand

   Open-source SDK from Browserbase for building browser agents with act, extract and observe calls in TypeScript, Python or Go.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Stagehand (official site)
6. 

   testRigor

   Write end-to-end tests for web, mobile, desktop and API in plain English, with a free public plan and a 14-day private trial.

   CodingAutomation and agents
   FreemiumVisit testRigor (official site)

agent-qa FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from vostride.com · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.