Auto-generated from skills/cover/SKILL.md

Source: https://github.com/yonatangross/orchestkit

Cover — Test Suite Generator

Generate comprehensive test suites for existing code with real-service integration testing and automated failure healing.

Quick Start

/ork:cover authentication flow
/ork:cover --model=opus payment processing
/ork:cover --tier=unit,integration user service
/ork:cover --real-services checkout pipeline

Argument Resolution

SCOPE = "$ARGUMENTS"  # e.g., "authentication flow"

# Flag parsing
MODEL_OVERRIDE = None
TIERS = ["unit", "integration", "e2e"]  # default: all three
REAL_SERVICES = False

for token in "$ARGUMENTS".split():
    if token.startswith("--model="):
        MODEL_OVERRIDE = token.split("=", 1)[1]
        SCOPE = SCOPE.replace(token, "").strip()
    elif token.startswith("--tier="):
        TIERS = token.split("=", 1)[1].split(",")
        SCOPE = SCOPE.replace(token, "").strip()
    elif token == "--real-services":
        REAL_SERVICES = True
        SCOPE = SCOPE.replace(token, "").strip()

Step -0.5: Effort-Aware Coverage Scaling (CC 2.1.76)

Scale test generation depth based on /effort level:

Effort Level	Tiers Generated	Agents	Heal Iterations
low	Unit only	1 agent	1 max
medium	Unit + Integration	2 agents	2 max
high (default)	Unit + Integration + E2E	3 agents	3 max

Override: Explicit --tier= flag or user selection overrides /effort downscaling.

Step -1: MCP Probe + Resume Check

# Probe MCPs (parallel):
ToolSearch(query="select:mcp__memory__search_nodes")
ToolSearch(query="select:mcp__context7__resolve-library-id")

Write(".claude/chain/capabilities.json", {
  "memory": <true if found>,
  "context7": <true if found>,
  "skill": "cover",
  "timestamp": now()
})

# Resume check:
Read(".claude/chain/state.json")
# If exists and skill == "cover": resume from current_phase
# Otherwise: initialize state

Step 0: Scope & Tier Selection

AskUserQuestion(
  questions=[
    {
      "question": "What test tiers should I generate?",
      "header": "Test Tiers",
      "options": [
        {"label": "Full coverage (Recommended)", "description": "Unit + Integration (real services) + E2E", "markdown": "```\nFull Coverage\n─────────────\n  Unit            Integration       E2E\n  ┌─────────┐    ┌─────────────┐  ┌──────────┐\n  │ AAA     │    │ Real DB     │  │Playwright│\n  │ Mocks   │    │ Real APIs   │  │Page obj  │\n  │ Factory │    │ Testcontain │  │A11y      │\n  └─────────┘    └─────────────┘  └──────────┘\n  3 parallel test-generator agents\n```"},
        {"label": "Unit + Integration", "description": "Skip E2E, focus on logic and service boundaries", "markdown": "```\nUnit + Integration\n──────────────────\n  Unit tests for business logic\n  Integration tests at API boundaries\n  Real services if docker-compose found\n  Skip: browser automation\n```"},
        {"label": "Unit only", "description": "Fast isolated tests for business logic", "markdown": "```\nUnit Only (~2 min)\n──────────────────\n  AAA pattern tests\n  MSW/VCR mocking\n  Factory-based data\n  Coverage gap analysis\n  Skip: real services, browser\n```"},
        {"label": "Integration only", "description": "API boundary and real-service tests", "markdown": "```\nIntegration Only\n────────────────\n  API endpoint tests (Supertest/httpx)\n  Database tests (real or in-memory)\n  Contract tests (Pact)\n  Testcontainers if available\n```"},
        {"label": "E2E only", "description": "Playwright browser tests", "markdown": "```\nE2E Only\n────────\n  Playwright page objects\n  User flow tests\n  Visual regression\n  Accessibility (axe-core)\n```"}
      ],
      "multiSelect": false
    },
    {
      "question": "Healing strategy for failing tests?",
      "header": "Failure Handling",
      "options": [
        {"label": "Auto-heal (Recommended)", "description": "Fix failing tests up to 3 iterations"},
        {"label": "Generate only", "description": "Write tests, report failures, don't fix"},
        {"label": "Strict", "description": "All tests must pass or abort"}
      ],
      "multiSelect": false
    }
  ]
)

Override TIERS based on selection. Skip this step if --tier= flag was provided.

Task Management (MANDATORY)

TaskCreate(
  subject=f"Cover: {SCOPE}",
  description="Generate comprehensive test suite with real-service testing",
  activeForm=f"Generating tests for {SCOPE}"
)

# Subtasks per phase
TaskCreate(subject="Discover scope and detect frameworks", activeForm="Discovering test scope")
TaskCreate(subject="Analyze coverage gaps", activeForm="Analyzing coverage gaps")
TaskCreate(subject="Generate tests (parallel per tier)", activeForm="Generating tests")
TaskCreate(subject="Execute generated tests", activeForm="Running tests")
TaskCreate(subject="Heal failing tests", activeForm="Healing test failures")
TaskCreate(subject="Generate coverage report", activeForm="Generating report")

6-Phase Workflow

Phase	Activities	Output
1. Discovery	Detect frameworks, scan scope, find untested code	Framework map, file list
2. Coverage Analysis	Run existing tests, map gaps per tier	Coverage baseline, gap map
3. Generation	Parallel test-generator agents per tier	Test files created
4. Execution	Run all generated tests	Pass/fail results
5. Heal	Fix failures, re-run (max 3 iterations)	Green test suite
6. Report	Coverage delta, test count, summary	Coverage report

Phase Handoffs

After Phase	Handoff File	Key Outputs
1. Discovery	01-cover-discovery.json	Frameworks, scope files, tier plan
2. Analysis	02-cover-analysis.json	Baseline coverage, gap map
3. Generation	03-cover-generation.json	Files created, test count per tier
5. Heal	05-cover-healed.json	Final pass/fail, iterations used

Phase 1: Discovery

Detect the project's test infrastructure and scope the work.

# PARALLEL — all in ONE message:
# 1. Framework detection (hook handles this, but also scan manually)
Grep(pattern="vitest|jest|mocha|playwright|cypress", glob="package.json", output_mode="content")
Grep(pattern="pytest|unittest|hypothesis", glob="pyproject.toml", output_mode="content")
Grep(pattern="pytest|unittest|hypothesis", glob="requirements*.txt", output_mode="content")

# 2. Real-service infrastructure
Glob(pattern="**/docker-compose*.yml")
Glob(pattern="**/testcontainers*")
Grep(pattern="testcontainers", glob="**/package.json", output_mode="content")
Grep(pattern="testcontainers", glob="**/requirements*.txt", output_mode="content")

# 3. Existing test structure
Glob(pattern="**/tests/**/*.test.*")
Glob(pattern="**/tests/**/*.spec.*")
Glob(pattern="**/__tests__/**/*")
Glob(pattern="**/test_*.py")

# 4. Scope files (what to test)
# If SCOPE specified, find matching source files
Grep(pattern=SCOPE, output_mode="files_with_matches")

Real-service decision:

• docker-compose*.yml found → integration tests use real services
• testcontainers in deps → use testcontainers for isolated service instances
• Neither found + --real-services flag → error: "No docker-compose or testcontainers found. Install testcontainers or remove --real-services flag."
• Neither found, no flag → integration tests use mocks (MSW/VCR)

Load real-service detection details: Read("${CLAUDE_SKILL_DIR}/references/real-service-detection.md")

Phase 2: Coverage Analysis

Run existing tests and identify gaps.

# Detect and run coverage command
# TypeScript: npx vitest run --coverage --reporter=json
# Python: pytest --cov=<scope> --cov-report=json
# Go: go test -coverprofile=coverage.out ./...

# Parse coverage output to identify:
# 1. Files with 0% coverage (priority targets)
# 2. Files below threshold (default 70%)
# 3. Uncovered functions/methods
# 4. Untested edge cases (error paths, boundary conditions)

Output coverage baseline to user immediately (progressive output).

Phase 3: Generation (Parallel Agents)

Spawn test-generator agents per tier. Launch ALL in ONE message with run_in_background=true.

# Unit tests agent
if "unit" in TIERS:
    Agent(
        subagent_type="test-generator",
        prompt=f"""Generate unit tests for: {SCOPE}
        Coverage gaps: {gap_map.unit_gaps}
        Framework: {detected_framework}
        Existing tests: {existing_test_files}

        Focus on:
        - AAA pattern (Arrange-Act-Assert)
        - Parametrized tests for multiple inputs
        - MSW/VCR for HTTP mocking (never mock fetch directly)
        - Factory-based test data (FactoryBoy/faker-js)
        - Edge cases: empty input, errors, timeouts, boundary values
        - Target: 90%+ business logic coverage""",
        isolation="worktree",
        run_in_background=True,
        max_turns=50,
        model=MODEL_OVERRIDE
    )

# Integration tests agent
if "integration" in TIERS:
    Agent(
        subagent_type="test-generator",
        prompt=f"""Generate integration tests for: {SCOPE}
        Coverage gaps: {gap_map.integration_gaps}
        Framework: {detected_framework}
        Real services available: {real_service_infra}

        Focus on:
        - API endpoint tests (Supertest/httpx)
        - Database tests with {'real DB via testcontainers/docker-compose' if real_services else 'in-memory/mocked DB'}
        - Contract tests (Pact) for service boundaries
        - Zod/Pydantic schema validation at edges
        - Fresh state per test (transaction rollback or cleanup)
        - Target: all API endpoints and service boundaries""",
        isolation="worktree",
        run_in_background=True,
        max_turns=50,
        model=MODEL_OVERRIDE
    )

# E2E tests agent
if "e2e" in TIERS:
    Agent(
        subagent_type="test-generator",
        prompt=f"""Generate E2E tests for: {SCOPE}
        Framework: Playwright
        Routes/pages: {discovered_routes}

        Focus on:
        - Semantic locators (getByRole > getByLabel > getByTestId)
        - Page Object Model for complex pages
        - User flow tests (happy path + error paths)
        - Accessibility tests (axe-core WCAG 2.2 AA)
        - Visual regression (toHaveScreenshot)
        - No hardcoded waits (use auto-wait)""",
        isolation="worktree",
        run_in_background=True,
        max_turns=50,
        model=MODEL_OVERRIDE
    )

Output each agent's results as soon as it returns — don't wait for all agents. This lets users see generated tests incrementally.

Partial results (CC 2.1.76): If an agent is killed (timeout, context limit), its response is tagged [PARTIAL RESULT]. Include partial tests but flag them in Phase 4.

Phase 4: Execution

Run all generated tests and collect results.

# Run test commands per tier (PARALLEL if independent):
# Unit: npx vitest run tests/unit/ OR pytest tests/unit/
# Integration: npx vitest run tests/integration/ OR pytest tests/integration/
# E2E: npx playwright test

# Collect: pass count, fail count, error details, coverage delta

Phase 5: Heal Loop

Fix failing tests iteratively. Max 3 iterations to prevent infinite loops.

for iteration in range(3):
    if all_tests_pass:
        break

    # For each failing test:
    # 1. Read the test file and the source code it tests
    # 2. Analyze the failure (assertion error? import error? timeout?)
    # 3. Fix the test (not the source code — tests only)
    # 4. Re-run the fixed tests

    # Common fixes:
    # - Wrong assertions (expected value mismatch)
    # - Missing imports or setup
    # - Stale selectors in E2E tests
    # - Race conditions (add proper waits)
    # - Mock configuration errors

Load heal strategy details: Read("${CLAUDE_SKILL_DIR}/references/heal-loop-strategy.md")

Boundary: heal fixes TESTS, not source code. If a test fails because the source code has a bug, report it — don't silently fix production code.

Phase 6: Report

Generate coverage report with before/after comparison.

Coverage Report: {SCOPE}
═══════════════════════════

Baseline → After
────────────────
  Unit:        67.2% → 91.3% (+24.1%)
  Integration: 42.0% → 78.5% (+36.5%)
  E2E:          0.0% → 65.0% (+65.0%)
  Overall:     48.4% → 82.1% (+33.7%)

Tests Generated
───────────────
  Unit:        23 tests (18 pass, 5 healed)
  Integration: 12 tests (10 pass, 2 healed)
  E2E:          8 tests (8 pass)
  Total:       43 tests

Heal Iterations: 2/3

Files Created
─────────────
  tests/unit/services/test_auth.py
  tests/unit/services/test_payment.py
  tests/integration/api/test_users.py
  tests/integration/api/test_checkout.py
  tests/e2e/checkout.spec.ts
  tests/e2e/pages/CheckoutPage.ts

Real Services Used: PostgreSQL (testcontainers), Redis (docker-compose)

Remaining Gaps
──────────────
  - src/services/notification.ts (0% — no tests generated, out of scope)
  - src/utils/crypto.ts (45% — edge cases not covered)

Next Steps
──────────
  /ork:verify {SCOPE}    # Grade the implementation + tests
  /ork:commit             # Commit generated tests
  /loop 10m npm test -- --coverage  # Watch coverage while coding

Coverage Drift Monitor (CC 2.1.71)

Optionally schedule weekly coverage drift detection:

# Guard: Skip cron in headless/CI (CLAUDE_CODE_DISABLE_CRON)
# if env CLAUDE_CODE_DISABLE_CRON is set, run a single check instead
CronCreate(
  schedule="0 2 * * 0",
  prompt="Weekly coverage drift check for {SCOPE}: npm test -- --coverage.
    If coverage >= baseline → CronDelete.
    If coverage drops > 5% → alert with regression details and recommendation."
)

Key Principles

• Output limits (CC 2.1.77+): Opus 4.6 defaults to 64k output tokens (128k upper bound). For large test suites, chunk generation across multiple agent turns if output approaches the limit.
• Tests only — never modify production source code, only generate test files
• Real services when available — prefer testcontainers/docker-compose over mocks for integration tests because mock/prod divergence causes silent failures in production
• Parallel generation — spawn one test-generator agent per tier in ONE message
• Heal, don't loop forever — max 3 iterations, then report remaining failures
• Progressive output — show results as each agent completes
• Factory over fixtures — use FactoryBoy/faker-js for test data, not hardcoded values
• Mock at network level — MSW/VCR, never mock fetch/axios directly

Related Skills

• ork:implement — generates tests during implementation (Phase 5); use /ork:cover after for deeper coverage
• ork:verify — grades existing tests 0-10; chain: implement → cover → verify
• testing-unit / testing-integration / testing-e2e — knowledge skills loaded by test-generator agents
• ork:commit — commit generated test files

References

Load on demand with Read("${CLAUDE_SKILL_DIR}/references/<file>"):

File	Content
real-service-detection.md	Docker-compose/testcontainers detection, service startup, teardown
heal-loop-strategy.md	Failure classification, fix patterns, iteration budget
coverage-report-template.md	Report format, delta calculation, gap analysis

Version: 1.0.0 (March 2026) — Initial release