agentleFS
mavka-aimavka-ai/unit-tests-skills/skills/generate-test-cases/SKILL.md

generate-test-cases

Skill55 starsChanged 59 days agoMITView on GitHub

Use when the user asks to analyze code for test coverage, list what test cases are needed, or review testing strategy — WITHOUT generating actual test code.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not skip reading the dependency classes.

    line 18
  2. Very specific:

    Read dependencies: Follow imports to read DTOs, entities, enums, and other types referenced by the target (as specified in …

    line 28
  3. Very specific:

    Check for existing tests: Search for existing test classes covering this target (as specified in existing-test-awareness rule) — if found, …

    line 29
  4. Distinctive:

    Stop at the descriptions — the list itself is this skill's deliverable

    line 41
  5. Very specific:

    Test method name format: {testedMethod}_{givenState}_{expectedOutcome}

    line 66
  6. Distinctive:

    If the specified target does not exist, inform the user with the exact path you searched and ask for clarification.

    line 79
  7. Distinctive:

    If the target code is in a language without specific rules, apply only the general rules and inform the user.

    line 82
  8. Distinctive:

    If the existing test class already covers all identified behaviors, output a summary stating that coverage is complete.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Generate Test Cases Skill

You will analyze code and generate a list of test cases that should be written for a given method/class. This skill outputs test case descriptions only — it does NOT generate actual test code.

Target to analyze: $ARGUMENTS

Quality Standards

  • Take your time to analyze the code thoroughly before listing test cases.
  • Quality is more important than speed — read all relevant source files and rules carefully.
  • Do not skip reading the dependency classes. Understanding the full context produces better test cases.

Instructions

Step 1: Read Rules and Analyze Context

  1. Read the rules from ./rules/general/ directory (see Rules Reference below)
  2. Read the target source file/class/method specified above
  3. Read dependencies: Follow imports to read DTOs, entities, enums, and other types referenced by the target (as specified in code-context-analysis rule)
  4. Check for existing tests: Search for existing test classes covering this target (as specified in existing-test-awareness rule) — if found, read it fully and focus only on behaviors not yet covered

Step 2: Generate Test Cases

  1. Analyze ALL code branches, including:
    • Success paths
    • Error/exception paths
    • Validation logic
    • Private/protected methods called by the target
    • Security annotations (if present)
  2. Apply the INCLUDE/EXCLUDE rules strictly
  3. Output the list of test cases in the specified format
  4. Stop at the descriptions — the list itself is this skill's deliverable
  5. Close by naming the next step: generate-tests <target> takes this list as its plan and writes the code from it. Hand the list forward as it stands — the user can review or amend it first

Output Format

For each test case, provide:

## Test Cases for {ClassName}.{methodName}

### 1. {testMethodName}
- **Given:** {preconditions/input state}
- **When:** {action being tested}
- **Then:** {expected outcome}
- **Code branch:** {which code path this covers}

### 2. {testMethodName}
...

Naming Convention

Test method name format: {testedMethod}_{givenState}_{expectedOutcome}

Examples:

  • calculateTotal_validProducts_returnsSum
  • calculateTotal_emptyList_throwsIllegalArgumentException
  • getUser_unauthorized_returns401
  • getUser_forbidden_returns403

Troubleshooting

Target file not found

If the specified target does not exist, inform the user with the exact path you searched and ask for clarification.

Unsupported language

If the target code is in a language without specific rules, apply only the general rules and inform the user.

All behaviors already covered

If the existing test class already covers all identified behaviors, output a summary stating that coverage is complete. List what is already tested. Do not invent additional test cases to justify the analysis.


Example

User says: "/generate-test-cases src/main/java/com/example/service/OrderService.java"

Step 1: Agent reads rules, reads OrderService.java, reads OrderRequest.java,
        Order.java (dependencies), checks for existing OrderServiceTest.java.

Step 2: Agent outputs:

## Test Cases for OrderService.createOrder

### 1. createOrder_validRequest_savesAndReturnsOrder
- **Given:** Valid OrderRequest with productId "product-1" and quantity 5
- **When:** createOrder is called
- **Then:** Order is saved to repository and returned with generated ID
- **Code branch:** Success path

### 2. createOrder_nullProductId_throwsIllegalArgumentException
- **Given:** OrderRequest with null productId
- **When:** createOrder is called
- **Then:** IllegalArgumentException is thrown
- **Code branch:** Validation — productId null check
...

Rules Reference

CRITICAL: You MUST read and apply all rules from the following files before generating test cases:

Maintenance note: General rules in ./rules/general/ are shared with the generate-tests skill (which has copies in rules/tests/general/). When updating rules, keep both locations in sync.

General Rules (Always Apply)

  • ./rules/general/test-case-generation-strategy.md - INCLUDE/EXCLUDE criteria for test cases
  • ./rules/general/naming-conventions.md - Test naming format
  • ./rules/general/general-principles.md - Core testing principles
  • ./rules/general/what-makes-good-test.md - Clarity, Completeness, Conciseness, Resilience
  • ./rules/general/keep-tests-focused.md - One scenario per test
  • ./rules/general/test-behaviors-not-methods.md - Separate tests for behaviors
  • ./rules/general/prefer-public-apis.md - Test public APIs over private methods
  • ./rules/general/cleanly-create-test-data.md - Use helpers and builders for test data
  • ./rules/general/keep-cause-effect-clear.md - Effects follow causes immediately
  • ./rules/general/no-logic-in-tests.md - KISS > DRY, avoid logic in assertions
  • ./rules/general/technology-stack-detection.md - Detect language and framework
  • ./rules/general/verify-relevant-arguments-only.md - Only verify relevant mock arguments
  • ./rules/general/existing-test-awareness.md - Check for existing tests, avoid duplicates
  • ./rules/general/code-context-analysis.md - Read dependencies before analyzing

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.