An AI-powered autonomous testing assistant that generates unit, integration, and E2E tests, diagnoses flaky test failures, and heals broken selectors.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentic-test-engineer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agentic-test-engineer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Agentic Test Engineer?
The Agentic Test Engineer is a comprehensive AI-native testing assistant designed to automate end-to-end quality assurance workflows. By integrating directly into your development lifecycle, it acts as an autonomous QA engineer capable of writing unit tests, testing complex API contracts, and structuring end-to-end browser interactions across modern frameworks like Playwright, Cypress, and Selenium. This capability is one of the most robust quality-focused options available in the catalog of Openclaw Skills.
Beyond simple code generation, this skill actively maintains your test suites. It utilizes visual AI heuristics to identify shifted UI components, automatically self-heals broken CSS or XPath selectors, and performs root-cause analysis on frustrating flaky tests. By keeping test coverage metrics aligned with actual code modifications, the Agentic Test Engineer minimizes manual maintenance overhead and accelerates CI/CD pipelines.
Agentic Test Engineer Use Cases
- Rapid Test Suite Bootstrapping: Automatically generate comprehensive unit and integration test coverage for newly written modules or legacy codebases.
- Self-Healing Broken Selectors: Instantly repair E2E browser test cases (Playwright/Cypress) when CSS classes or DOM hierarchies change during UI updates.
- Flaky Test Troubleshooting: Diagnose intermittent CI/CD pipeline failures by analyzing timing issues, async race conditions, and network states.
- Targeted Coverage Enhancement: Scan code repositories to pinpoint untested logical branches and generate specific edge-case inputs to close the coverage gap.
How Agentic Test Engineer Works
- Context Detection: The assistant analyzes your project framework (Python, Node.js, Go, Java) and target testing layer (unit, integration, API, or E2E browser automation) to understand the requirements.
- Test Generation & Structuring: Based on input source code, API specs, or user stories, it builds fully structured test suites including happy-path validations, edge cases, negative scenarios, and mock data setups.
- Visual Heuristic Self-Healing: When a front-end test breaks, the visual engine analyzes the changed interface to generate resilient alternative selectors utilizing semantic roles and accessibility labels.
- Flakiness Diagnosis: The agent inspects historical test failures and timing behaviors to identify race conditions or unhandled promises, suggesting bulletproof waiting strategies.
- Coverage Evaluation: It parses existing codebase execution files, measures exact coverage metrics, and outputs actionable recommendations to secure uncovered logical pathways.
Agentic Test Engineer Setup
To utilize the Agentic Test Engineer as part of your collection of Openclaw Skills, you must declare it in your agent configurations. Ensure you have the corresponding testing engines installed locally or in your CI environment for compiling generated tests.
Install Necessary Testing Frameworks
Ensure your local runtime environment is equipped with the framework you wish to output. For example, for Playwright:
npm install -D @playwright/test
npx playwright install
Accessing the Skill
To prompt the agent to start generating or repairing tests, ensure your workspace is mounted so it can scan source code files. Invoke the tool with clear contextual commands such as:
openclaw run 'generate tests for src/utils/math.ts'
Agentic Test Engineer Data Schema & Taxonomy
Supported Frameworks and Engines
| Level | Supported Ecosystems | Key Implementations |
|---|---|---|
| Unit / Integration | Python, JavaScript, TypeScript, Java, Go | pytest, Jest, Vitest, JUnit, testify |
| E2E / Browser | Playwright, Cypress, Selenium | Page Object Model, Visual AI Element Locating |
| API / Contract | REST, GraphQL | OpenAPI Specs, Pact, Postman Collections |
Input and Output Data Formats
- Input Triggers: Supports plain-text diagnostic queries, failing test logs, UI snapshots, and target source code files.
- Output Test Files: Generates production-ready test scripts with mock fixtures, parameterized data, and robust wait conditions.
- Diagnostic Reports: Provides flaky test probability matrices and semantic repair alternative suggestions mapped to semantic accessibility parameters.
Agentic Test Engineer Advanced Features
- Semantic Role Binding: Prioritizes robust accessibility labels (like 'getByRole') over volatile CSS classes for bulletproof self-healing.
- Visual Heuristic Locator Engine: Uses proximity matching and text-content strategies to automatically pinpoint moved elements on updated screens.
- Asynchronous Retry Modeling: Wraps tricky asynchronous UI updates in explicit poll configurations to eliminate race conditions in E2E environments.
- Multi-Agent Collaborative Pipeline: Seamlessly pairs with architectural planning skills to plan, write, and audit massive test suites autonomously.
SKILL.md
Loading
Related Openclaw Skills
Excel / XLSX
A high-reliability AI coding skill designed to create, inspect, and edit Microsoft Excel workbooks, preserving complex formulas, formatting, and template structures.
Analytical Thinking
A structured, framework-driven skill that enables AI agents to decompose complex problems, test hypotheses, map evidence, and generate clear, synthesized decisions.
Creative Thinking
A structured, agentic framework designed to separate divergent generation from convergent evaluation to produce high-quality, novel solutions.
Ethical Thinking
A structured ethical analysis skill for AI agents to map stakeholders, navigate value tensions, assess power asymmetries, and recommend balanced actions.
AI Meeting Intelligence Expert
Streamline your meeting lifecycle by automatically turning raw text, transcriptions, and notes into structured action items, summaries, and email follow-ups.
Agent Skills Framework Explorer
An AI-powered assistant for exploring, comparing, and implementing major AI agent skills frameworks, SDKs, and Model Context Protocol (MCP) integrations.
Featured*
