# Actionbook Web Test

Run browser-based web tests against websites using Actionbook CLI.

- Repository: https://github.com/actionbook/actionbook (1623 stars)
- Category: Design
- License: Apache-2.0
- Last updated: 2026-10-19
- Runs scripts: yes, review before use

## Install

Claude Code: `npx skills add actionbook/actionbook --skill actionbook-web-test -g -a claude-code -y`
Codex: `npx skills add actionbook/actionbook --skill actionbook-web-test -g -a codex -y`
Cursor: `npx skills add actionbook/actionbook --skill actionbook-web-test -g -a cursor -y`
Gemini CLI: `npx skills add actionbook/actionbook --skill actionbook-web-test -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add actionbook/actionbook`

## Description

Run browser-based web tests against websites using Actionbook CLI. Activate when the user wants to test a website workflow, run smoke tests, verify a user flow, check if a web application works, run regression tests, or validate browser-based interactions. Supports test definition, execution, assertion, reporting, and json-ui visual report generation.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

## When to Use This Skill

Activate when the user:

- Asks to "test", "verify", "check", or "validate" a website workflow
- Wants to run smoke tests or health checks on a web application
- Needs to verify a user flow works end-to-end (login, checkout, search, etc.)
- Asks to "run regression tests" or "does this still work?"
- Wants to confirm a deployment didn't break functionality
- Needs to monitor a website's functionality on a schedule
- Builds browser-based test suites without writing Playwright/Cypress code

## What actionbook-web-test Provides

actionbook-web-test transforms web tests from coded test scripts into **declarative YAML workflows** executed by AI agents via Actionbook CLI.

| Benefit | How |
|---------|-----|
| **AI-native recovery** | When a selector fails, the agent snapshots the live page and finds the equivalent element |
| **Actionbook-managed selectors** | Pre-verified selectors with health scores, no manual maintenance |
| **Cross-project reusability** | YAML workflows work anywhere Actionbook CLI is installed |
| **No test framework required** | No Playwright/Cypress/Jest setup, just `actionbook browser` commands |
| **Human-readable tests** | YAML workflows are readable by non-developers |
| **Visual test reports** | json-ui powered HTML reports with metrics, step details, and failure screenshots |

## Test Workflow Format

Tests are defined as YAML files in a `tests/` directory. Each file describes one test workflow.

```yaml
name: example-test
description: What this test verifies
url: https://example.com
tags: [smoke, critical]
timeout: 30000  # ms, default 30000

# Pre-fetch verified selectors from Actionbook
actions:
  - "example.com:/:default"

# Environment variables (support {{env.VAR}} templates)
env:
  USERNAME: "test-user"
  PASSWORD: "{{env.TEST_PASSWORD}}"

# Browser setup options
setup:
  headless: true
  auto_dismiss_dialogs: true
  no_animations: true

The rest: https://github.com/actionbook/actionbook/blob/HEAD/playground/actionbook-web-test/SKILL.md