qa-agent 开源项目推荐（16 个）

MigoXLab /webqa-agent

自主网页浏览器代理，审计工程师和氛围编码创作者的性能、功能和用户体验。网站自主评估测试 Agent，支持 GUI/CLI 一键完成性能、功能使用与交互体验的测试评估

srmorete /mobile-device-mcp

An MCP server to use with iOS and Android. Seamlessly maps Native screens and WebViews. Multiple parallel devices supported.

rakutentech /sorify

Sorify —— 面向 AI 智能体的 AI 智能体化 QA 平台。E2E AI 测试——完整的用户管理、测试运行器、CI webhook、调度器、通知，以及用于测试运行、创建和管理的 MCP。附带 Chrome 扩展和用于录制事件以供 AI 编写测试的 MCP。

Lykhoyda /QaReN

A QA engineer for your coding agent. It tests each React Native change on a simulator and tells you PASS or FAIL, with video proof.

Anasss /qa-orchestra

AI-powered QA orchestration for Claude Code. 10 agents, from diff analysis to bug reports.

Vip3r-MC /Veta

Autonomous AI agent swarm for visual, functional, and accessibility testing of Android apps and mobile web - powered by vision-capable AI models and containerized Android instances.

Manancode /qa-agent

AI QA agent that converts Notion requirements into Playwright test runs, posts Slack approve/dismiss alerts, and performs post-deploy production validation via Amplitude event streams

pugazhmukilan /QA-Agent-OceanAI

autonomous QA agent capable of constructing a “testing brain” from project documentation. The system will ingest support documents (e.g., product specifications, UI/UX guidelines, mock APIs) alongside the HTML structure of a target web project. Using these inputs, the agent should:

dreamquality /cursor-qa-memory-bank

🔧 Comprehensive QA Test Automation Framework with memory persistence and structured 5-mode workflow system for complete testing lifecycle management

BAKUGOS1 /QaAgent

Local-first TypeScript + Playwright QA agent for browser automation, evidence capture, safety checks, and Excel-first website QA reports.

Kalaiselvan-21 /OpenSource-AITestCaseGenerator

Self Experimenting Kalai - AI Test Case Generator powered by Ollama LLM and FAISS vector database. Completely free of cost. No OpenAPI Key is required

MobiDev-Org /treegress-browser-core

Treegress Browser Core for Treegress Browser MCP

ManishJoc14 /QAAgent

This is a modular, agentic QA engine that can automatically test websites or URLs using AI and registered tools. It leverages an LLM provider (currently Mistral) and allows tool-based automation, like browser actions, bash commands, and more.

dooperdrew /ghost-qa

GhostQA 👻: Autonomous, invisible Windows Desktop & Hardware-in-the-Loop (HIL) QA Agent. Tests hands-free in the background without stealing focus, correlates hardware logs, and critiques visual UX with Multimodal AI.

2xgrowthagency /make-no-mistakes

Risk-proportionate Make No Mistakes workflow: conditional builder checks, independent Crosscheck QA, and live E2E evidence.

srmorete /aguara

AI Agent for Mobile QA. Write plain language scenarios, run them on iOS and Android, get structured step-by-step report with screenshots as evidence.