AGENT SKILL

test-patterns

Generate tests following project conventions with proper patterns for unit, integration, and e2e testing

License 许可
MIT
最近更新
2026-10-03
Stars
★ 0
Forks
0

01 前言 Front Matter

name
test-patterns
license
MIT
metadata
{ "audience": "developers", "workflow": "general" }
description
Generate tests following project conventions with proper patterns for unit, integration, and e2e testing
compatibility
opencode

03 SKILL.md 正文

What I do

  • Analyze existing test patterns in the project
  • Generate tests that follow the same style and conventions
  • Cover happy paths, edge cases, and error scenarios
  • Use the project's testing framework and utilities

When to use me

Use this skill when you need to write tests for new or existing code. This skill first analyzes existing tests to match the project's style.

Process

  1. Discover: Find existing tests to understand patterns

    • Look for *.test.*, *.spec.*, __tests__/ directories
    • Identify testing framework (Jest, Vitest, Pytest, JUnit, etc.)
    • Note assertion style, mock patterns, and setup/teardown
  2. Analyze: Read the code under test

    • Identify public API surface
    • Find branching logic and edge cases
    • Note dependencies that need mocking
  3. Generate: Write tests following discovered patterns

Test Patterns by Framework

JavaScript/TypeScript (Vitest/Jest)

describe("ModuleName", () => {
  // Setup
  beforeEach(() => { /* ... */ })

  describe("methodName", () => {
    it("should return expected value for valid input", () => {
      // Arrange
      const input = createTestInput()
      // Act
      const result = methodName(input)
      // Assert
      expect(result).toEqual(expectedValue)
    })

    it("should throw for invalid input", () => {
      expect(() => methodName(null)).toThrow(ExpectedError)
    })
  })
})

Python (Pytest)

class TestModuleName:
    @pytest.fixture
    def setup(self):
        return create_test_fixture()

    def test_method_returns_expected_for_valid_input(self, setup):
        result = method_name(setup)
        assert result == expected_value

    def test_method_raises_for_invalid_input(self):
        with pytest.raises(ExpectedError):
            method_name(None)

Java (JUnit 5)

@DisplayName("ModuleName Tests")
class ModuleNameTest {
    @BeforeEach
    void setUp() { /* ... */ }

    @Test
    @DisplayName("should return expected value for valid input")
    void methodName_validInput_returnsExpected() {
        var result = moduleName.methodName(validInput);
        assertEquals(expected, result);
    }

    @Test
    @DisplayName("should throw for invalid input")
    void methodName_invalidInput_throws() {
        assertThrows(ExpectedException.class,
            () -> moduleName.methodName(null));
    }
}

Coverage Targets

  • Unit Tests: Every public method, all branches
  • Edge Cases: Null/empty, boundary values, large inputs
  • Error Paths: Invalid input, network failures, timeouts
  • Integration: Key workflows between components

04 同仓库更多 Skills

balagnanas
balagnanas/LocalMachineAISetup

git-release

Create consistent releases with changelogs, version bumps, and release notes from git history

★ 0
balagnanas
balagnanas/LocalMachineAISetup

dependency-audit

Scan dependencies for CVEs, outdated packages, license issues, and unused deps to produce a prioritized remediation list

★ 0
balagnanas
balagnanas/LocalMachineAISetup

ci-pipeline

Generate CI/CD pipeline config for detected platform with lint, test, build, and deploy stages

★ 0
balagnanas
balagnanas/LocalMachineAISetup

changelog-generate

Generate CHANGELOG.md from git history in Keep a Changelog format with Added, Changed, Deprecated, Removed, Fixed, and Security categories

★ 0
balagnanas
balagnanas/LocalMachineAISetup

list-models

List the models available from the OpenCode Zen (provider ID `opencode`, API base https://opencode.ai/zen/v1) and OpenCode Go (provider ID `opencode-go`, API base https://opencode.ai/zen/go/v1) providers. Runs `opencode models opencode` / `opencode models opencode-go` for plain id lists or `--verbose` for per-model metadata (display name, cost, context window, capabilities). Use when the user asks what models are available in OpenCode Zen or OpenCode Go, wants to browse `opencode/...` or `opencode-go/...` model ids, or needs model cost, context-size, or capability details.

★ 0
balagnanas
balagnanas/LocalMachineAISetup

list-models

List the models available from the OpenCode Zen (provider ID `opencode`, API base https://opencode.ai/zen/v1) and OpenCode Go (provider ID `opencode-go`, API base https://opencode.ai/zen/go/v1) providers. Runs `opencode models opencode` / `opencode models opencode-go` for plain id lists or `--verbose` for per-model metadata (display name, cost, context window, capabilities). Use when the user asks what models are available in OpenCode Zen or OpenCode Go, wants to browse `opencode/...` or `opencode-go/...` model ids, or needs model cost, context-size, or capability details.

★ 0