
webqa-agent
Autonomous web testing agent that checks functionality, performance, UX and security
- Pricing
- Open source
- Platforms
- CLI, Web, self-hosted
- GitHub stars
- 240
- Last commit
- 5 months ago
- Language
- Python
- License
- Apache-2.0
About webqa-agent
webqa-agent is an open-source agent for browser automation. It is available on CLI, Web, self-hosted and API.
WebQA Agent drops the test script: you state a goal in one sentence and it drives a browser to try it, using multimodal understanding of the page. The README describes three forms of testing, a lightweight Flash mode for quick smoke tests, a Standard Generate mode that discovers pages and plans tests (focused when you give goals, exploratory when you do not), and a Run mode that executes YAML instructions for repeatable regression checks.
It can be used from a CLI, a web dashboard, or from coding agents such as Cursor and Claude Code through MCP or a skill. For teams, a Docker Compose deployment adds a dashboard with test management and history; you supply an LLM API key in the environment file.
It is a Python project under Apache-2.0, with a Discord and WeChat community linked from the README.
What it does
- Flash mode
- Turns a one-sentence goal into a quick browser test.
- Generate mode
- Discovers the site and plans tests, either focused on goals or exploratory.
- Run mode
- Executes YAML steps with expected checks for regression testing.
- MCP and skill access
- Lets Cursor, Claude Code and similar agents trigger tests.
- Docker Compose dashboard
- Adds test management and run history for teams.
Good for
- Smoke-test a new feature from inside an IDE
- Explore a freshly built site for broken flows
- Re-run a fixed regression checklist before release
- Audit pages for performance and security issues
Pricing
Open source
Apache-2.0; needs your own LLM API key
webqa-agent FAQ
Do I need to write test scripts?
No, Flash and Generate modes work from a natural-language goal.
How do I run it with a dashboard?
Copy the example .env, add your LLM API key and run the start script in deploy/docker-compose.
Can coding agents call it?
Yes, through MCP or a skill, as the README describes.
Is webqa-agent free?
Yes. webqa-agent is open source, so you can use and self-host it for free.
What are the best alternatives to webqa-agent?
The closest alternatives to webqa-agent are deepseek-harness, gemini-cli, learn-claude-code, Codewhale and oh-my-pi.
Alternatives to webqa-agent

deepseek-harness
AgentOpen-source agent harness from DeepSeek, built on a plugin architecture
- Open source
- Open-source agent
- 282K GitHub stars
- Multi-agent frameworks

gemini-cli
AgentOpen-source terminal agent for Gemini Code Assist workflows on Google Cloud
- Open source
- Open-source agent
- 109K GitHub stars
- AI models and LLM tooling

learn-claude-code
AgentHands-on course teaching how to build the harness around an AI coding agent
- Open source
- Hosted agent
- 80.5K GitHub stars
- AI models and LLM tooling

Codewhale
AgentTerminal coding agent in Rust that works with any model and can split tasks across agents
- Open source
- Open-source agent
- 41.2K GitHub stars
- Coding

oh-my-pi
AgentTerminal coding agent with LSP, debugger access, subagents and a native Rust core
- Open source
- Open-source agent
- 39.2K GitHub stars
- Coding

openclaude
AgentTerminal coding agent that talks to OpenAI, Gemini, Ollama and 200-plus other models
- Open source
- Open-source agent
- 34.4K GitHub stars
- Coding