Test Strategy Design Skill

Purpose

Enable the QA Lead Agent to define the overall approach to quality assurance for a project, ensuring all risks are mitigated through appropriate testing.

Test Pyramid Strategy

Based on Mike Cohn's Test Pyramid:

1. Unit Tests (70%): Component level, fast, isolated. (Dev responsibility)
2. Integration Tests (20%): API level, service interaction. (Dev/QA responsibility)
3. E2E / UI Tests (10%): User journey, slow, fragile. (QA responsibility)

Test Plan Components

1. Scope

• In Scope: Modules, browsers (Chrome, Safari), devices (iOS, Android).
• Out of Scope: Performance (unless specified), third-party system internals.

2. Testing Types

• Functional: Sanity, Smoke, Regression.
• Non-Functional: Performance, Security, Accessibility, Usability.

3. Environment Strategy

• Dev: Unstable, for unit tests.
• QA/Staging: Stable, mirror of Prod, for functional/regression.
• Prod: For smoke tests / monitoring.

4. Data Strategy

• Synthetic data generation?
• Anonymized prod dump?
• Hardcoded test users?

Automation Framework Design

Selection Criteria

• Tech Stack: Playwright (JS/TS), Selenium (Java/Python), Cypress (JS).
• CI/CD Integration: Github Actions, Jenkins.
• Reporting: Allure, HTML reports.

Best Practices

• Page Object Model (POM): Separation of page structure from tests.
• Atomic Tests: Each test is independent.
• Data Driven: Separate data from logic.

Risk-Based Testing

Prioritize testing based on:

• Impact: What happens if this fails? (Financial loss, data loss?)
• Probability: How likely is it to fail? (New complex code vs. old stable code).

Deliverables

• Master Test Plan (MTP).
• Test Case Suite.
• Defect Reports.
• Test Summary Report.