test-coverage-improver

name	test-coverage-improver
description	テストカバレッジを分析し、最もインパクトの大きいテスト追加を提案する。 「テスト改善」「カバレッジ」「test-coverage」で呼び出される。 モジュールごとのカバレッジ分析、未テストのエッジケース検出、 テストスケルトンの自動生成を行う。
metadata	{"short-description":"テストカバレッジ改善提案"}

/test-coverage-improver -- Test Coverage Analysis & Improvement

$ARGUMENTS (optional: specific module or file to analyze)

Overview

Analyzes existing tests, identifies coverage gaps, and suggests the most impactful tests to add. Generates test skeletons for quick implementation.

Workflow

Step 1: Test Inventory

• Scan tests/ directory for all test files
• Count test cases per module/hook
• Build a coverage map: which source files have tests, which don't

Step 2: Gap Analysis

• Identify source files with NO corresponding tests
• For tested files, check coverage of:
  • Happy path (normal input, expected output)
  • Boundary values (min, max, empty, zero)
  • Error cases (invalid input, error conditions)
  • Edge cases (null, empty string, special characters)
• Prioritize gaps by risk level (security hooks > utility modules)

Step 3: Impact Ranking

Rank suggested tests by impact:

Priority	Criteria
HIGH	Security-related hooks (gate, injection-guard) without edge case tests
HIGH	Core modules (session, stage) with missing error path tests
MEDIUM	Utility modules with partial coverage
LOW	Already well-tested modules needing only edge cases

Step 4: Skeleton Generation

For each HIGH priority gap, generate a test skeleton:

const { describe, it } = require("node:test");
const assert = require("node:assert/strict");

describe("{module} - {scenario}", () => {
  it("should {expected behavior}", () => {
    // Arrange
    // Act
    // Assert
    assert.ok(false, "TODO: implement");
  });
});

Step 5: Report

Output:

1. Coverage summary table (module / test count / coverage level / priority)
2. Top 5 recommended tests to add (with skeletons)
3. Estimated effort (LOW / MEDIUM / HIGH per test)

Notes

• Uses node:test runner (shield-harness standard)
• Test naming convention: test_{subject}_{condition}_{expected_result}
• Run tests with: node --test --test-concurrency=1 tests/
• This skill generates suggestions and skeletons -- actual implementation is manual or via /tdd