CI/CD Pipeline Skill

When to Apply: Reference when setting up automated test execution in CI/CD pipelines, configuring GitHub Actions workflows for test automation, integrating Jenkins/GitLab CI with cloud testing, or implementing quality gates and notifications. Triggers on "CI/CD", "pipeline", "GitHub Actions", "Jenkins", "GitLab CI", "Azure DevOps".

Rule Categories by Priority:

Quick Reference:

1. Pipeline Configuration (CRITICAL) pipeline-github-actionsUse actions/checkout@v4, actions/setup-node@v4, artifact upload pipeline-jenkins — Jenkinsfile with parallel stages (unit + E2E + cloud) pipeline-gitlab — .gitlab-ci.yml with parallel matrix and JUnit reports pipeline-azure — azure-pipelines.yml with variable groups
2. Secrets Management (CRITICAL) secrets-githubStore LT_USERNAME/LT_ACCESS_KEY in Settings → Secrets secrets-jenkins — Use credentials store with credentials() binding secrets-gitlab — Store in Settings → CI/CD → Variables
3. Quality Gates (MEDIUM) gate-coverageEnforce minimum code coverage thresholds gate-test-results — Fail pipeline on test failures gate-artifact-upload — Always upload test results as artifacts (if: always())

Playbook Deep Dive (10 sections): §1 GitHub Actions — Production Workflows · §2 Caching & Optimization · §3 GitLab CI — Production Pipeline · §4 Jenkins Pipeline · §5 Quality Gates & Checks · §6 Secrets & Environment Management · §7 Docker Compose for Test Services · §8 Notification & Reporting · §9 Debugging Table · §10 Best Practices

The agent loads from 1 on-demand reference files (playbook.md) containing detailed implementation guides, cloud integration patterns, debugging tables, and language-specific best practices.

Shared Infrastructure

TestMu AI Cloud Reference

URL:

Universal cloud reference shared across all 46 skills covering authentication, WebSocket/Hub endpoints, desktop browsers (Chrome, Edge, Firefox, Safari, pw-chromium, pw-firefox, pw-webkit), desktop platforms (Windows 11/10, macOS Sequoia through Catalina), Android devices (Pixel 8/7/6/5, Galaxy S24/S23/S22, OnePlus 11/10 Pro + 100 more), iOS devices (iPhone 16/15/14, iPad Pro/Air), LT:Options capability reference, test status reporting in 5 languages, Lambda Tunnel for localhost, geo-location testing (50+ countries), and network/video capture.

Evaluation Test Cases

URL:

Each skill has eval JSON files with test cases defining queries, expected behaviors, and categories (basic-generation, cloud-routing, language-detection, pom-structure, mobile, debugging, negative-trigger, scaffold).

Validation & Scripts

URL:

validate_skills.py validates all skills against quality standards (YAML frontmatter, line counts, reference files, categories).

How to Use

Install a Single Skill

git clone https://github.com/LambdaTest/agent-skills.git
cp -r agent-skills/playwright-skill .claude/skills/

Install All Skills

# Claude Code
git clone https://github.com/LambdaTest/agent-skills.git .claude/skills/agent-skills

# Cursor
git clone https://github.com/LambdaTest/agent-skills.git .cursor/skills/agent-skills

# Gemini CLI
git clone https://github.com/LambdaTest/agent-skills.git .gemini/skills/agent-skills

Read Individual Skill Files

Each skill directory contains:

• SKILL.mdCore instructions with decision trees and quick patterns
• reference/playbook.mdComplete implementation guide with all code examples
• reference/advanced-patterns.mdAdvanced topics and edge cases
• reference/cloud-integration.mdTestMu AI cloud-specific patterns (for E2E/mobile skills)

Validate Skills

python3 scripts/validate_skills.py