AI自动化测试
AI Skill Hub 推荐使用：AI自动化测试 是一款优质的Agent工作流。AI 综合评分 7.5 分，在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案，这是一个值得深入了解的选择。
📚 深度解析
AI自动化测试 工作流的设计遵循"最小配置，最大复用"原则：核心逻辑已经封装好，用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制，在网络波动或 API 限速等情况下仍能稳定运行，适合作为生产环境的自动化基础设施。
在实际部署时，建议先在测试环境中运行 3-5 次，验证各个环节的输出结果符合预期，再部署到生产环境。AI Skill Hub 评分 7.5 分，是同类 Agent 工作流中的精选推荐。
📋 工具概览
将英文测试规格转换为自愈Playwright测试
AI自动化测试 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
📖 中文文档
将英文测试规格转换为自愈Playwright测试
AI自动化测试 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
# 方式一：pip 安装（推荐）
pip install quorvex_ai
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install quorvex_ai
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/NihadMemmedli/quorvex_ai
cd quorvex_ai
pip install -e .
# 验证安装
python -c "import quorvex_ai; print('安装成功')"
- 访问 GitHub 仓库获取工作流文件
- 在对应平台（Dify / Flowise / Make 等）中找到「导入工作流」功能
- 上传工作流文件
- 按照提示配置必要的环境变量和 API Key
- 运行测试确认流程正常后投入使用
# 命令行使用
quorvex_ai --help
# 基本用法
quorvex_ai input_file -o output_file
# Python 代码中调用
import quorvex_ai
# 示例
result = quorvex_ai.process("input")
print(result)
# quorvex_ai 配置文件示例（config.yml） app: name: "quorvex_ai" debug: false log_level: "INFO" # 运行时指定配置文件 quorvex_ai --config config.yml # 或通过环境变量配置 export QUORVEX_AI_API_KEY="your-key" export QUORVEX_AI_OUTPUT_DIR="./output"
简介
<p align="center"> <h1 align="center">Quorvex AI</h1> <p align="center"> <strong>Self-hosted AI testing agents that turn specs into validated Playwright tests.</strong> </p> <p align="center"> Generate code you can inspect, commit, and run in CI without runtime AI dependency. </p> <p align="center"> <a href="https://github.com/NihadMemmedli/quorvex_ai/stargazers"><img src="https://img.shields.io/github/stars/NihadMemmedli/quorvex_ai?style=social" alt="GitHub Stars"></a> <a href="https://github.com/NihadMemmedli/quorvex_ai/network/members"><img src="https://img.shields.io/github/forks/NihadMemmedli/quorvex_ai?style=social" alt="GitHub Forks"></a> </p> <p align="center"> <a href="https://github.com/NihadMemmedli/quorvex_ai/actions/workflows/ci.yml"><img src="https://github.com/NihadMemmedli/quorvex_ai/actions/workflows/ci.yml/badge.svg" alt="CI"></a> <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT"></a> <a href="https://www.python.org/downloads/"><img src="https://img.shields.io/badge/Python-3.10+-3776AB.svg?logo=python&logoColor=white" alt="Python 3.10+"></a> <a href="https://nodejs.org/"><img src="https://img.shields.io/badge/Node.js-20+-339933.svg?logo=nodedotjs&logoColor=white" alt="Node.js 20+"></a> <a href="https://playwright.dev/"><img src="https://img.shields.io/badge/Playwright-45ba4b.svg?logo=playwright&logoColor=white" alt="Playwright"></a> <a href="https://fastapi.tiangolo.com/"><img src="https://img.shields.io/badge/FastAPI-009688.svg?logo=fastapi&logoColor=white" alt="FastAPI"></a> <a href="https://nextjs.org/"><img src="https://img.shields.io/badge/Next.js-black.svg?logo=next.js&logoColor=white" alt="Next.js"></a> <a href="https://github.com/NihadMemmedli/quorvex_ai/commits/main"><img src="https://img.shields.io/github/last-commit/NihadMemmedli/quorvex_ai" alt="Last Commit"></a> <a href="CONTRIBUTING.md"><img src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg" alt="PRs Welcome"></a> </p> <p align="center"> <a href="https://nihadmemmedli.github.io/quorvex_ai/"><strong>Documentation</strong></a> • <a href="https://nihadmemmedli.github.io/quorvex_ai/tutorials/getting-started/">Getting Started</a> • <a href="https://github.com/NihadMemmedli/quorvex_ai/issues">Issues</a> • <a href="CONTRIBUTING.md">Contributing</a> </p> </p>
---
Quorvex AI is for teams that already know Playwright is the right test runtime, but do not want to hand-write every brittle end-to-end flow. Describe a user flow, import a PRD or OpenAPI file, or let an agent explore a real app; Quorvex plans the flow, generates Playwright TypeScript, validates it in a browser, and repairs selector or timing failures when it can.
The output is normal code your team owns: inspect it, commit it, and run it in CI with no runtime AI dependency. Around that core workflow, Quorvex also supports PRD-to-tests, API checks, K6 load tests, security scans, database quality checks, mobile smoke flows, LLM evaluation suites, CI quality gates, and autonomous coverage discovery.
Plain-English specs become validated test code through planning, browser execution, and healing.
Description
Verify user can log in with valid credentials.
Description
Verify a user can complete the checkout process.
Features
| Area | What Quorvex AI supports |
|---|---|
| Test generation | Plain-English specs to Playwright, native validation, Smart Check reuse, hybrid healing, visual regression, reusable @include templates |
| AutoPilot & agents | App discovery, live browser state, generated task artifacts, custom agent definitions, persistent autonomous missions, recurring or long-running operation, approval gates |
| Requirements & coverage | PRD upload, feature extraction, requirements generation, duplicate detection, RTM, coverage gaps, suggested tests |
| Specialized testing | OpenAPI/API testing, K6 load testing, quick/Nuclei/ZAP security scans, database quality checks, LLM evaluation, Appium mobile smoke flows |
| Quality intelligence | Regression batches, flaky test detection, pass-rate trends, failure classification, analytics, generated reports |
| Integrations | GitHub Actions, GitLab CI, PR advisor, quality gates, workflow PRs, TestRail sync, Jira issue creation |
| Operations | Project isolation, RBAC, encrypted credentials, schedules, browser pools, Redis queues, MinIO storage, backup, archival, Docker/Swarm/Kubernetes assets |
---
Prerequisites
| Requirement | Version | Notes |
|---|---|---|
| Docker | 20+ | Required for recommended setup |
| Docker Compose | 2.x | Included with Docker Desktop |
| Git | 2.x | For cloning the repository |
Which Setup Should You Use?
| Setup | Best for | Database | Command |
|---|---|---|---|
| Minimal Docker | Fast first trial on smaller machines | SQLite | docker compose -f docker-compose.minimal.yml up -d |
| Full Docker dev | Complete self-hosted evaluation with dashboard, queues, storage, VNC, and security scanning | PostgreSQL | make prod-dev |
| Local native | Contributors and backend/frontend development | PostgreSQL if Docker is running, otherwise SQLite | make setup && make dev |
| Production | Hardened single-host deployment | PostgreSQL | make prod-up |
Installation
```bash
Docker mode (recommended)
cp .env.prod.example .env.prod
AI Provider Setup
Quorvex AI requires an Anthropic-compatible API provider. Three options are supported:
Option 1: Z.ai GLM Coding Plan (Recommended Default)
QUORVEX_LLM_PROVIDER=anthropic_compatible
QUORVEX_LLM_API_KEY=your-z-ai-token
QUORVEX_LLM_BASE_URL=https://api.z.ai/api/anthropic
QUORVEX_LLM_LIGHT_MODEL=glm-4.5-air
QUORVEX_LLM_STANDARD_MODEL=glm-5-turbo
QUORVEX_LLM_DEEP_MODEL=glm-5.1
QUORVEX_LLM_TOOL_DEEP_MODEL=glm-5.1
QUORVEX_LLM_CHAT_MODEL=glm-5-turbo
API_TIMEOUT_MS=3000000Create a key in the Z.ai API Keys flow. Quorvex mirrors canonical QUORVEX_LLM_* settings to Anthropic-compatible aliases for Claude Code and SDK clients.
Option 2: OpenRouter (Free Models Available)
OpenRouter provides access to free and paid LLM models through an Anthropic-compatible API.
QUORVEX_LLM_PROVIDER=anthropic_compatible
QUORVEX_LLM_API_KEY=sk-or-v1-your-openrouter-key
QUORVEX_LLM_BASE_URL=https://openrouter.ai/api
QUORVEX_LLM_STANDARD_MODEL=meta-llama/llama-3.2-3b-instruct:freePopular free models on OpenRouter:
| Model | Provider | Context | Best For |
|---|---|---|---|
meta-llama/llama-3.2-3b-instruct:free | Meta | 131k | General tasks |
google/gemini-2.0-flash-exp:free | 1M | Fast responses | |
qwen/qwen-2.5-7b-instruct:free | Alibaba | 32k | Coding assistance |
Free models have rate limits. For production use, consider paid models or Z.ai/Anthropic direct.
Option 3: Anthropic Direct
QUORVEX_LLM_PROVIDER=anthropic_compatible
QUORVEX_LLM_API_KEY=sk-ant-your-api-key
QUORVEX_LLM_BASE_URL=https://api.anthropic.com
QUORVEX_LLM_STANDARD_MODEL=claude-sonnet-4-20250514Sign up at console.anthropic.com to get an API key.
Docker Mode (Recommended)
```bash cp .env.prod.example .env.prod
Docker mode (recommended)
make prod-dev # Start all services with local code mounting make prod-restart # Restart backend (picks up code changes) make prod-logs # Tail production logs make prod-status # Show status of all services make prod-down # Stop all services make prod-build # Rebuild Docker images
Quick Start
Product Screenshots
| Dashboard | API testing | Workflow monitor |
|---|---|---|
|  |  |  |
| AutoPilot | Test runs | Settings |
|---|---|---|
|  |  |  |
---
2. Configure your AI provider credentials
cp .env.prod.example .env.prod
Edit .env.prod and set QUORVEX_LLM_API_KEY (see Configuration below)
Optional: confirm local and production env files are readable
make check-env
Activate the virtual environment
source venv/bin/activate
Configuration
Configuration depends on your running mode:
```bash
Optional Configuration
```env
Edit .env.prod, then validate it
make check-env make prod-dev ```
- All services in Docker containers (backend, frontend, PostgreSQL, Redis, MinIO, VNC)
- Local
./orchestratorand./web/srcmounted for hot-reload - Backend API on http://localhost:8001, frontend on http://localhost:3000, API docs on http://localhost:8001/docs
- VNC browser view on http://localhost:6080 and MinIO console on http://localhost:9001
Edit .env, then validate it
make check-env make dev # Start backend + frontend natively ```
- Starts PostgreSQL via Docker if available, falls back to SQLite
- Hot-reload enabled for backend and frontend
CLI Mode (No dashboard)
source venv/bin/activate
python orchestrator/cli.py specs/your-test.md- Direct command-line execution, no database required
- Artifacts stored in
runs/TIMESTAMP/ - Useful for CI/CD pipelines
---
Run a spec through the native pipeline
python orchestrator/cli.py specs/your-test.md
Pipeline Modes
Native Pipeline (Default)
The recommended pipeline. AI agents use a real browser at every stage for maximum reliability.
python orchestrator/cli.py specs/your-test.mdPRD Pipeline
Convert a PDF product requirements document into test specs and then into tests.
python orchestrator/cli.py your-prd.pdf --prd
python orchestrator/cli.py your-prd.pdf --prd --feature "User Login"How Quorvex AI Compares
| Capability | Quorvex AI | Shortest | Octomind | testRigor | Playwright Test Agents |
|---|---|---|---|---|---|
| Natural-language authoring | ✅ Specs, PRDs, chat, exploration | ✅ | ✅ Prompt/discovery | ✅ Plain English | ✅ Agent prompts |
| Standard Playwright code | ✅ Owned repo code | Runtime-oriented Playwright | ✅ Portable Playwright | ❌ Proprietary no-code runtime | ✅ Generated tests |
| Generate once, run natively | ✅ | ❌ AI used during execution | ✅ Cloud/local execution | Managed platform runtime | ✅ |
| Self-healing / repair | ✅ Native, hybrid, standard | Not advertised | ✅ Source-level healing | ✅ AI maintenance | ✅ Healer agent |
| Web QA dashboard | ✅ Full platform | Not advertised | ✅ Hosted QA dashboard | ✅ Hosted QA dashboard | ❌ Editor/agent workflow |
| Requirements, RTM, coverage | ✅ Built in | Not advertised | Not advertised | Not advertised | Plan/spec files only |
| PRD to tests | ✅ Upload + feature workspace | Not advertised | Not advertised | Not advertised | PRD context for agents |
| Autonomous missions | ✅ Scheduled/approval-gated | Not advertised | Not advertised | Not advertised | Agent loop, not platform missions |
| API testing | ✅ OpenAPI import + API specs | ✅ Natural-language API tests | Not advertised as API testing | ✅ API commands | Via code/MCP |
| Load testing | ✅ K6 workers/results | Not advertised | Not advertised | Not advertised | Via custom code |
| Security testing | ✅ ZAP + Nuclei | Not advertised | Not advertised | Not advertised | Via custom code |
| Database testing | ✅ Connections, schema checks | Via callback code | Not advertised | ✅ Database query support | Via custom code |
| LLM evaluation | ✅ Providers, datasets, comparisons | Not advertised | Not advertised | Not advertised | Via custom code |
| CI/CD and PR advisor | ✅ GitHub/GitLab + quality gates | ✅ CI headless runs | ✅ CI/CD | ✅ CI integrations | ✅ In repo workflows |
| Test management integrations | ✅ TestRail + Jira | Not advertised | TestRail on higher tiers | Integrations advertised | Via custom code |
| Self-hosted / private deployment | ✅ Full stack | ✅ Package/repo | Hosted SaaS + private workers | Hosted SaaS | ✅ Local repo agents |
| Open source / license | ✅ MIT | ✅ MIT | ❌ Commercial | ❌ Commercial | ✅ Playwright |
Generate once, run forever -- Unlike runtime-first AI runners, Quorvex AI outputs stable Playwright code. Subsequent runs execute natively with zero AI cost. Comparison notes are based on public product documentation and repositories checked on May 23, 2026. "Not advertised" means the capability was not clearly documented as a first-class product feature, not that it is impossible to build with custom code.
Detailed comparisons | Why we built this
---
Troubleshooting
| Symptom | Solution |
|---|---|
| "QUORVEX_LLM_API_KEY not set" | Check .env file, run make check-env |
| "Database connection refused" | Run docker compose up -d db or use SQLite (default) |
| Generated test selector fails | Self-healer auto-fixes; use --hybrid for complex cases |
| "No target URL found in spec" | Spec must contain a URL (e.g., "Navigate to https://...") |
| Test timeout on complex pages | Use --hybrid or increase exploration depth |
| "Module not found" errors | Re-run make setup to reinstall dependencies |
For more diagnostics, see the Troubleshooting Guide.
---
Quorvex AI 是一个自主的 AI 测试代理，能够将规范转换为验证的 Playwright 测试。它可以生成代码供开发者检查、提交和在 CI 中运行，而无需在运行时依赖 AI。
Quorvex AI 支持以下功能：测试生成、AutoPilot 和代理、App 发现、直播浏览器状态、生成任务艺术品、自定义代理定义、持久的自主任务、重复或长时间运行的操作、批准门户等。
Quorvex AI 需要以下环境依赖和系统要求：Docker 20+、Docker Compose 2.x、Git 2.x 等。
Quorvex AI 支持以下安装方式：最小 Docker、全 Docker 开发环境、native 本地环境等。具体安装步骤如下：
Quorvex AI 的快速启动步骤包括：配置 AI 提供商凭证、编辑 .env.prod 文件、设置 QUORVEX_LLM_API_KEY 等。
Quorvex AI 的配置说明包括：配置 AI 提供商凭证、编辑 .env.prod 文件、设置 QUORVEX_LLM_API_KEY 等。
Quorvex AI 的 CLI 模式（无仪表板）可以直接在命令行执行，无需数据库支持，生成的艺术品存储在 runs/TIMESTAMP/ 中，适合用于 CI/CD 管道等。
Quorvex AI 的工作流包括：运行规范通过 native 管道、Pipeline 模式（包括 Native Pipeline 和 Hybrid Pipeline 等）等。
Quorvex AI 的常见问题包括：QUORVEX_LLM_API_KEY 未设置、数据库连接拒绝等，解决方案包括检查 .env 文件、运行 make check-env 等。
自动化测试工具，提高测试效率
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
- 构建企业知识库 / RAG 检索应用的团队
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- 生产部署优先使用 Docker Compose 隔离依赖，并挂载 volume 持久化数据
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- 容器内无法访问宿主机 localhost — 使用 host.docker.internal
- Python 依赖冲突：建议用 venv / uv 隔离环境
- Docker：quorvex_ai 提供官方镜像，docker compose up 一键启动
- CLI：直接 npm install -g / pip install，命令行调用
- 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
⚡ 核心功能
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
- 构建企业知识库 / RAG 检索应用的团队
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- 生产部署优先使用 Docker Compose 隔离依赖，并挂载 volume 持久化数据
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- 容器内无法访问宿主机 localhost — 使用 host.docker.internal
- Python 依赖冲突：建议用 venv / uv 隔离环境
👥 适合人群
🎯 使用场景
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
⚖️ 优点与不足
- +MIT 协议，可免费商用
- +大幅减少重复性人工操作
- +可视化流程，清晰直观
- +可扩展性强，支持复杂场景
- −初始配置和调试需投入一定时间
- −强依赖外部服务的稳定性
- −复杂场景需具备一定技术基础
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一，可自由商用、修改、分发，仅需保留版权声明。
🔗 相关工具推荐
❓ 常见问题 FAQ
总体来看，AI自动化测试 是一款质量良好的Agent工作流，在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态，建议收藏备用，结合自身场景选择合适时机引入使用。
| 原始名称 | quorvex_ai |
| 原始描述 | 开源AI工作流：Convert plain English test specs into self-healing Playwright tests using AI. Br。⭐34 · Python |
| Topics | ai测试自动化 |
| GitHub | https://github.com/NihadMemmedli/quorvex_ai |
| License | MIT |
| 语言 | Python |
收录时间：2026-06-05 · 更新时间：2026-06-05 · License：MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。