software_test_case_design

This skill should be used when generating test cases, writing test cases, designing test cases, supplementing exception scenarios and boundary values, designing complex interaction test cases, outputting standardized test cases, compatibility testing, adaptation testing, API testing, security testing, UI visual testing, mobile/App/mini-program/H5/desktop/PC Web testing, AI Agent testing, linkage testing, or routing testing. Only focuses on writing test cases, does not involve test plans, test strategies, or test automation scripts.

Execution Flow

1. Identify test case type:
   • Contains "API Test"/"API Testing" → API test case
   • Contains "Agent Test"/"Agent"/"AI Agent" → Agent test case
   • Default → Functional test case
2. Load capability files: Always load templates/common-rules.md (rules). Select capability files by type:
   • API test → core-capabilities/api-testing.md
   • Agent test → core-capabilities/agent-testing.md + Part 1: Test Case Design Methods and Part 2: Test Case Quality Standards from core-capabilities/functional-testing.md
   • Functional test → core-capabilities/functional-testing.md
3. Load platform-specific files (skip this step for API tests):
   • Match platform keywords → load platform/{platform}.md
4. Generate test cases: Generate test cases according to the loaded capability files and platform-specific files
5. Self-check:
   • API test → checklists/api-checklist.md
   • Agent test → checklists/agent-checklist.md + Section I: Functional Testing Checklist from checklists/common-checklist.md
   • Agent test + platform → checklists/agent-checklist.md + Section I: Functional Testing Checklist from checklists/common-checklist.md + checklists/{platform}-checklist.md
   • Functional test + platform → checklists/common-checklist.md + checklists/{platform}-checklist.md
   • Functional test without platform → checklists/common-checklist.md
6. Output: Output in the Markdown table format specified by examples/format-spec.md

Capability Boundaries

✅ Can generate: Functional testing, API testing, AI Agent testing, basic security testing, platform-specific testing ❌ Cannot generate: Test plans, test strategies, test plans, penetration test execution, vulnerability scanning, performance/stress testing (concurrent/stress/load), automation scripts

Instruction Mapping Table

The rules file templates/common-rules.md is always loaded. Capability files are selected by test case type. Platform files are overlaid as needed.

Capability Files

Keyword Trigger	Load	Notes
"API Test", "API Testing"	core-capabilities/api-testing.md	Used independently, no platform overlay
"Agent Test", "Agent", "AI Agent"	core-capabilities/agent-testing.md + Part 1: Test Case Design Methods and Part 2: Test Case Quality Standards from core-capabilities/functional-testing.md	Can overlay any platform
Default	core-capabilities/functional-testing.md	Functional testing

Platform Files (overlay on top of capability files, except for API tests)

Keyword Trigger	Load	Notes
"Mobile Testing", "App Testing"	platform/mobile-app.md	Gestures/interruptions/network/permissions/push/compatibility/performance
"Mini-Program Testing"	platform/mini-program.md	Lifecycle/authorization/sharing/payment/navigation/subscription
"Mobile Web Testing", "H5 Testing"	platform/mobile-web.md	Responsive/touch/browser/viewport/H5/SEO
"Desktop Testing", "Desktop App Testing"	platform/desktop.md	Window/shortcuts/files/system integration/multi-monitor
"PC Web Testing", "Web Testing"	platform/pc-web.md	Browser/layout/keyboard/forms/session/routing/drag-and-drop