Prompt file imported from traefikturkey/onramp (.github/prompts/check.prompt.md). Copyright stays with the author.

Quality Check and Fix

Name: check
Author: traefikturkey

Objective

Run comprehensive quality gates and systematically address any failures until all checks pass.

Context

This is the primary quality gate before committing code. All checks must pass.

Priority: Tests green > Lint clean > Types valid

Workflow

Step 1: Run Tests

Python (uv):

uv run pytest -q

Python (traditional):

pytest

Node.js:

npm test

Result:

✅ All tests pass → Proceed to Step 2
❌ Tests fail → Fix and re-run until green

Step 2: Run Linting (Optional)

Python (Ruff):

uv run ruff check .

Python (flake8):

uv run flake8 src/ tests/

JavaScript/TypeScript:

npm run lint

Result:

✅ No lint errors → Proceed to Step 3
❌ Lint errors → Fix and re-run

Step 3: Run Type Checking (Optional)

Python (mypy):

uv run mypy src/

TypeScript:

npm run type-check

Result:

✅ No type errors → All checks passed!
❌ Type errors → Fix and re-run

Step 4: Run Makefile Target (If Available)

If Makefile has a check target:

make check

This typically runs all quality checks in sequence.

Fixing Failures

Test Failures

Strategy:

Read error messages carefully
Identify the root cause
Make minimal, targeted fixes
Re-run tests to verify fix
Repeat until all tests pass

Common issues:

Import errors → Check __init__.py files, update imports
Missing dependencies → uv add --dev <package>
Deleted files → Update tests or remove obsolete test files
Changed APIs → Update test expectations

Example fix:

# Error: ModuleNotFoundError: No module named 'requests'
uv add requests

# Re-run tests
uv run pytest -q

Lint Failures

Strategy:

Auto-fix what's safe: uv run ruff check --fix .
Manually fix remaining issues
Re-run linter to verify

Common issues:

Unused imports → Remove them
Line too long → Break into multiple lines
Missing docstrings → Add brief docstrings
Code style → Run formatter: uv run ruff format .

Type Errors

Strategy:

Add missing type hints
Fix incorrect type annotations
Use # type: ignore sparingly for genuinely complex cases
Prefer fixing the code over ignoring errors

Common issues:

Missing type hints → Add them
Incorrect types → Fix annotations
Incompatible types → Adjust code or types

Constraints

From Repository Instructions

✅ Use uv run pytest for uv projects (never call pytest directly in uv projects)
✅ Make small, surgical fixes that preserve behavior
✅ Only add dependencies when explicitly required by failures
❌ Never add unnecessary flags: uv run pytest (not uv run -m pytest)
❌ Never prefix commands with cd
❌ Never add || true to hide failures

Best Practices

Prefer small changes that preserve existing behavior
Fix root causes rather than symptoms
Test frequently - run tests after each fix
Keep commits clean - all checks should pass before committing
Ask for clarification if the fix is unclear or risky

Example Sessions

Example 1: Tests pass, lint has minor issues

# Step 1: Run tests
uv run pytest -q
# ✅ All tests pass

# Step 2: Run linter
uv run ruff check .
# ❌ Found 3 errors:
#   - Unused import 'sys' in src/main.py
#   - Line too long in src/utils.py
#   - Missing docstring in src/helpers.py

# Fix: Auto-fix safe issues
uv run ruff check --fix .
# ✅ Fixed 1 issue automatically

# Fix: Manually fix line length
# [Edit src/utils.py to break long line]

# Fix: Add docstring
# [Edit src/helpers.py to add docstring]

# Re-run linter
uv run ruff check .
# ✅ No errors found

# Report: All checks passed!

Example 2: Test failures due to missing dependency

# Step 1: Run tests
uv run pytest -q
# ❌ Tests failed:
#   ModuleNotFoundError: No module named 'pydantic'

# Fix: Add missing dependency
uv add pydantic

# Re-run tests
uv run pytest -q
# ✅ All tests pass

# Step 2: Run linter
uv run ruff check .
# ✅ No errors found

# Report: All checks passed!

Example 3: Complex test failure requiring code fix

# Step 1: Run tests
uv run pytest -q
# ❌ Tests failed:
#   test_user_service.py::test_create_user FAILED
#   AssertionError: Expected status 201, got 400

# Analyze: Check test and implementation
# [Read test_user_service.py]
# [Read user_service.py]

# Fix: Update validation logic in user_service.py
# [Edit user_service.py to fix validation]

# Re-run tests
uv run pytest -q
# ✅ All tests pass

# Continue with linting...

Output Format

Success Report

✅ Quality checks passed!

Tests: All passing
Lint: No errors
Types: All valid

Ready to commit.

Failure Report

❌ Quality checks failed

Tests: 2 failing
  - test_auth.py::test_login - AssertionError
  - test_user.py::test_create - ValidationError

Fixes applied:
  - Added missing pydantic dependency
  - Updated test expectations

Re-running tests...

Makefile Integration

If `make check` exists:

Use it as the final comprehensive gate after individual checks pass:

# Run individual checks first
uv run pytest -q  # ✅
uv run ruff check .  # ✅

# Final comprehensive check
make check  # ✅

# Report: All checks passed!

Makefile check target example:

.PHONY: check
check: test lint type-check
	@echo "All checks passed!"

Troubleshooting

Tests hanging or taking too long

# Run with timeout
uv run pytest --timeout=60

# Run specific tests
uv run pytest tests/unit/ -k "fast"

# Skip slow tests
uv run pytest -m "not slow"

Linter too strict

# Check what can be auto-fixed
uv run ruff check --fix .

# For specific rules to ignore, update pyproject.toml

Type checker too strict

# Run with less strict checking
uv run mypy src/ --no-strict-optional

# Or add type: ignore comments strategically

Customization Notes

CUSTOMIZE THIS PROMPT for your project:

Test command: Update for your test framework
Linter: ruff, flake8, eslint, etc.
Type checker: mypy, pyright, tsc
Makefile: Adjust if using different targets
Quality gates: Add/remove checks as needed
CI alignment: Match your CI/CD pipeline checks

Source: Consolidated from agent-spike, onboard, attempt-one

# Quality Check and Fix ## Objective Run comprehensive quality gates and systematically address any failures until all checks pass. ## Context This is the **primary quality gate** before committing code. All checks must pass. **Priority:** Tests green > Lint clean > Types valid --- ## Workflow ### Step 1: Run Tests **Python (uv):** ```bash uv run pytest -q ``` **Python (traditional):** ```bash pytest ``` **Node.js:** ```bash npm test ``` **Result:** - ✅ All tests pass → Proceed to Step 2 - ❌ Tests fail → Fix and re-run until green ### Step 2: Run Linting (Optional) **Python (Ruff):** ```bash uv run ruff check . ``` **Python (flake8):** ```bash uv run flake8 src/ tests/ ``` **JavaScript/TypeScript:** ```bash npm run lint ``` **Result:** - ✅ No lint errors → Proceed to Step 3 - ❌ Lint errors → Fix and re-run ### Step 3: Run Type Checking (Optional) **Python (mypy):** ```bash uv run mypy src/ ``` **TypeScript:** ```bash npm run type-check ``` **Result:** - ✅ No type errors → All checks passed! - ❌ Type errors → Fix and re-run ### Step 4: Run Makefile Target (If Available) **If Makefile has a `check` target:** ```bash make check ``` This typically runs all quality checks in sequence. --- ## Fixing Failures ### Test Failures **Strategy:** 1. Read error messages carefully 2. Identify the root cause 3. Make minimal, targeted fixes 4. Re-run tests to verify fix 5. Repeat until all tests pass **Common issues:** - Import errors → Check `__init__.py` files, update imports - Missing dependencies → `uv add --dev <package>` - Deleted files → Update tests or remove obsolete test files - Changed APIs → Update test expectations **Example fix:** ```bash # Error: ModuleNotFoundError: No module named 'requests' uv add requests # Re-run tests uv run pytest -q ``` ### Lint Failures **Strategy:** 1. Auto-fix what's safe: `uv run ruff check --fix .` 2. Manually fix remaining issues 3. Re-run linter to verify **Common issues:** - Unused imports → Remove them - Line too long → Break into multiple lines - Missing docstrings → Add brief docstrings - Code style → Run formatter: `uv run ruff format .` ### Type Errors **Strategy:** 1. Add missing type hints 2. Fix incorrect type annotations 3. Use `# type: ignore` sparingly for genuinely complex cases 4. Prefer fixing the code over ignoring errors **Common issues:** - Missing type hints → Add them - Incorrect types → Fix annotations - Incompatible types → Adjust code or types --- ## Constraints ### From Repository Instructions - ✅ Use `uv run pytest` for uv projects (never call `pytest` directly in uv projects) - ✅ Make small, surgical fixes that preserve behavior - ✅ Only add dependencies when explicitly required by failures - ❌ Never add unnecessary flags: `uv run pytest` (not `uv run -m pytest`) - ❌ Never prefix commands with `cd` - ❌ Never add `|| true` to hide failures ### Best Practices - **Prefer small changes** that preserve existing behavior - **Fix root causes** rather than symptoms - **Test frequently** - run tests after each fix - **Keep commits clean** - all checks should pass before committing - **Ask for clarification** if the fix is unclear or risky --- ## Example Sessions ### Example 1: Tests pass, lint has minor issues ```bash # Step 1: Run tests uv run pytest -q # ✅ All tests pass # Step 2: Run linter uv run ruff check . # ❌ Found 3 errors: # - Unused import 'sys' in src/main.py # - Line too long in src/utils.py # - Missing docstring in src/helpers.py # Fix: Auto-fix safe issues uv run ruff check --fix . # ✅ Fixed 1 issue automatically # Fix: Manually fix line length # [Edit src/utils.py to break long line] # Fix: Add docstring # [Edit src/helpers.py to add docstring] # Re-run linter uv run ruff check . # ✅ No errors found # Report: All checks passed! ``` ### Example 2: Test failures due to missing dependency ```bash # Step 1: Run tests uv run pytest -q # ❌ Tests failed: # ModuleNotFoundError: No module named 'pydantic' # Fix: Add missing dependency uv add pydantic # Re-run tests uv run pytest -q # ✅ All tests pass # Step 2: Run linter uv run ruff check . # ✅ No errors found # Report: All checks passed! ``` ### Example 3: Complex test failure requiring code fix ```bash # Step 1: Run tests uv run pytest -q # ❌ Tests failed: # test_user_service.py::test_create_user FAILED # AssertionError: Expected status 201, got 400 # Analyze: Check test and implementation # [Read test_user_service.py] # [Read user_service.py] # Fix: Update validation logic in user_service.py # [Edit user_service.py to fix validation] # Re-run tests uv run pytest -q # ✅ All tests pass # Continue with linting... ``` --- ## Output Format ### Success Report ``` ✅ Quality checks passed! Tests: All passing Lint: No errors Types: All valid Ready to commit. ``` ### Failure Report ``` ❌ Quality checks failed Tests: 2 failing - test_auth.py::test_login - AssertionError - test_user.py::test_create - ValidationError Fixes applied: - Added missing pydantic dependency - Updated test expectations Re-running tests... ``` --- ## Makefile Integration ### If `make check` exists: Use it as the **final comprehensive gate** after individual checks pass: ```bash # Run individual checks first uv run pytest -q # ✅ uv run ruff check . # ✅ # Final comprehensive check make check # ✅ # Report: All checks passed! ``` ### Makefile check target example: ```makefile .PHONY: check check: test lint type-check @echo "All checks passed!" ``` --- ## Troubleshooting ### Tests hanging or taking too long ```bash # Run with timeout uv run pytest --timeout=60 # Run specific tests uv run pytest tests/unit/ -k "fast" # Skip slow tests uv run pytest -m "not slow" ``` ### Linter too strict ```bash # Check what can be auto-fixed uv run ruff check --fix . # For specific rules to ignore, update pyproject.toml ``` ### Type checker too strict ```bash # Run with less strict checking uv run mypy src/ --no-strict-optional # Or add type: ignore comments strategically ``` --- ## Customization Notes **CUSTOMIZE THIS PROMPT** for your project: 1. **Test command:** Update for your test framework 2. **Linter:** ruff, flake8, eslint, etc. 3. **Type checker:** mypy, pyright, tsc 4. **Makefile:** Adjust if using different targets 5. **Quality gates:** Add/remove checks as needed 6. **CI alignment:** Match your CI/CD pipeline checks **Source:** Consolidated from agent-spike, onboard, attempt-one

{ "prompt": "# Quality Check and Fix\n\n## Objective\n\nRun comprehensive quality gates and systematically address any failures until all checks pass.\n\n## Context\n\nThis is the **primary quality gate** before committing code. All checks must pass.\n\n**Priority:** Tests green > Lint clean > Types valid\n\n---\n\n## Workflow\n\n### Step 1: Run Tests\n\n**Python (uv):**\n```bash\nuv run pytest -q\n```\n\n**Python (traditional):**\n```bash\npytest\n```\n\n**Node.js:**\n```bash\nnpm test\n```\n\n**Result:**\n- ✅ All tests pass → Proceed to Step 2\n- ❌ Tests fail → Fix and re-run until green\n\n### Step 2: Run Linting (Optional)\n\n**Python (Ruff):**\n```bash\nuv run ruff check .\n```\n\n**Python (flake8):**\n```bash\nuv run flake8 src/ tests/\n```\n\n**JavaScript/TypeScript:**\n```bash\nnpm run lint\n```\n\n**Result:**\n- ✅ No lint errors → Proceed to Step 3\n- ❌ Lint errors → Fix and re-run\n\n### Step 3: Run Type Checking (Optional)\n\n**Python (mypy):**\n```bash\nuv run mypy src/\n```\n\n**TypeScript:**\n```bash\nnpm run type-check\n```\n\n**Result:**\n- ✅ No type errors → All checks passed!\n- ❌ Type errors → Fix and re-run\n\n### Step 4: Run Makefile Target (If Available)\n\n**If Makefile has a `check` target:**\n```bash\nmake check\n```\n\nThis typically runs all quality checks in sequence.\n\n---\n\n## Fixing Failures\n\n### Test Failures\n\n**Strategy:**\n1. Read error messages carefully\n2. Identify the root cause\n3. Make minimal, targeted fixes\n4. Re-run tests to verify fix\n5. Repeat until all tests pass\n\n**Common issues:**\n- Import errors → Check `__init__.py` files, update imports\n- Missing dependencies → `uv add --dev <package>`\n- Deleted files → Update tests or remove obsolete test files\n- Changed APIs → Update test expectations\n\n**Example fix:**\n```bash\n# Error: ModuleNotFoundError: No module named 'requests'\nuv add requests\n\n# Re-run tests\nuv run pytest -q\n```\n\n### Lint Failures\n\n**Strategy:**\n1. Auto-fix what's safe: `uv run ruff check --fix .`\n2. Manually fix remaining issues\n3. Re-run linter to verify\n\n**Common issues:**\n- Unused imports → Remove them\n- Line too long → Break into multiple lines\n- Missing docstrings → Add brief docstrings\n- Code style → Run formatter: `uv run ruff format .`\n\n### Type Errors\n\n**Strategy:**\n1. Add missing type hints\n2. Fix incorrect type annotations\n3. Use `# type: ignore` sparingly for genuinely complex cases\n4. Prefer fixing the code over ignoring errors\n\n**Common issues:**\n- Missing type hints → Add them\n- Incorrect types → Fix annotations\n- Incompatible types → Adjust code or types\n\n---\n\n## Constraints\n\n### From Repository Instructions\n\n- ✅ Use `uv run pytest` for uv projects (never call `pytest` directly in uv projects)\n- ✅ Make small, surgical fixes that preserve behavior\n- ✅ Only add dependencies when explicitly required by failures\n- ❌ Never add unnecessary flags: `uv run pytest` (not `uv run -m pytest`)\n- ❌ Never prefix commands with `cd`\n- ❌ Never add `|| true` to hide failures\n\n### Best Practices\n\n- **Prefer small changes** that preserve existing behavior\n- **Fix root causes** rather than symptoms\n- **Test frequently** - run tests after each fix\n- **Keep commits clean** - all checks should pass before committing\n- **Ask for clarification** if the fix is unclear or risky\n\n---\n\n## Example Sessions\n\n### Example 1: Tests pass, lint has minor issues\n\n```bash\n# Step 1: Run tests\nuv run pytest -q\n# ✅ All tests pass\n\n# Step 2: Run linter\nuv run ruff check .\n# ❌ Found 3 errors:\n# - Unused import 'sys' in src/main.py\n# - Line too long in src/utils.py\n# - Missing docstring in src/helpers.py\n\n# Fix: Auto-fix safe issues\nuv run ruff check --fix .\n# ✅ Fixed 1 issue automatically\n\n# Fix: Manually fix line length\n# [Edit src/utils.py to break long line]\n\n# Fix: Add docstring\n# [Edit src/helpers.py to add docstring]\n\n# Re-run linter\nuv run ruff check .\n# ✅ No errors found\n\n# Report: All checks passed!\n```\n\n### Example 2: Test failures due to missing dependency\n\n```bash\n# Step 1: Run tests\nuv run pytest -q\n# ❌ Tests failed:\n# ModuleNotFoundError: No module named 'pydantic'\n\n# Fix: Add missing dependency\nuv add pydantic\n\n# Re-run tests\nuv run pytest -q\n# ✅ All tests pass\n\n# Step 2: Run linter\nuv run ruff check .\n# ✅ No errors found\n\n# Report: All checks passed!\n```\n\n### Example 3: Complex test failure requiring code fix\n\n```bash\n# Step 1: Run tests\nuv run pytest -q\n# ❌ Tests failed:\n# test_user_service.py::test_create_user FAILED\n# AssertionError: Expected status 201, got 400\n\n# Analyze: Check test and implementation\n# [Read test_user_service.py]\n# [Read user_service.py]\n\n# Fix: Update validation logic in user_service.py\n# [Edit user_service.py to fix validation]\n\n# Re-run tests\nuv run pytest -q\n# ✅ All tests pass\n\n# Continue with linting...\n```\n\n---\n\n## Output Format\n\n### Success Report\n\n```\n✅ Quality checks passed!\n\nTests: All passing\nLint: No errors\nTypes: All valid\n\nReady to commit.\n```\n\n### Failure Report\n\n```\n❌ Quality checks failed\n\nTests: 2 failing\n - test_auth.py::test_login - AssertionError\n - test_user.py::test_create - ValidationError\n\nFixes applied:\n - Added missing pydantic dependency\n - Updated test expectations\n\nRe-running tests...\n```\n\n---\n\n## Makefile Integration\n\n### If `make check` exists:\n\nUse it as the **final comprehensive gate** after individual checks pass:\n\n```bash\n# Run individual checks first\nuv run pytest -q # ✅\nuv run ruff check . # ✅\n\n# Final comprehensive check\nmake check # ✅\n\n# Report: All checks passed!\n```\n\n### Makefile check target example:\n\n```makefile\n.PHONY: check\ncheck: test lint type-check\n\t@echo \"All checks passed!\"\n```\n\n---\n\n## Troubleshooting\n\n### Tests hanging or taking too long\n\n```bash\n# Run with timeout\nuv run pytest --timeout=60\n\n# Run specific tests\nuv run pytest tests/unit/ -k \"fast\"\n\n# Skip slow tests\nuv run pytest -m \"not slow\"\n```\n\n### Linter too strict\n\n```bash\n# Check what can be auto-fixed\nuv run ruff check --fix .\n\n# For specific rules to ignore, update pyproject.toml\n```\n\n### Type checker too strict\n\n```bash\n# Run with less strict checking\nuv run mypy src/ --no-strict-optional\n\n# Or add type: ignore comments strategically\n```\n\n---\n\n## Customization Notes\n\n**CUSTOMIZE THIS PROMPT** for your project:\n\n1. **Test command:** Update for your test framework\n2. **Linter:** ruff, flake8, eslint, etc.\n3. **Type checker:** mypy, pyright, tsc\n4. **Makefile:** Adjust if using different targets\n5. **Quality gates:** Add/remove checks as needed\n6. **CI alignment:** Match your CI/CD pipeline checks\n\n**Source:** Consolidated from agent-spike, onboard, attempt-one", "variables": [], "notes": "Run comprehensive quality checks: tests, linting, type checking. Fix failures systematically. Mode: agent.", "metadata": { "source": { "provider": "github-prompts", "repository": "https://github.com/traefikturkey/onramp", "path": ".github/prompts/check.prompt.md", "ref": "4f5c540d2d4cc89ddc031270ca3e07d8af014424", "url": "https://github.com/traefikturkey/onramp/blob/4f5c540d2d4cc89ddc031270ca3e07d8af014424/.github/prompts/check.prompt.md", "key": "traefikturkey/onramp/.github/prompts/check.prompt.md" } }, "model_hints": [ "gpt-5-mini" ] }

check

About

Quality Check and Fix

Objective

Context

Workflow

Step 1: Run Tests

Step 2: Run Linting (Optional)

Step 3: Run Type Checking (Optional)

Step 4: Run Makefile Target (If Available)

Fixing Failures

Test Failures

Lint Failures

Type Errors

Constraints

From Repository Instructions

Best Practices

Example Sessions

Example 1: Tests pass, lint has minor issues

Example 2: Test failures due to missing dependency

Example 3: Complex test failure requiring code fix

Output Format

Success Report

Failure Report

Makefile Integration

If `make check` exists:

Makefile check target example:

Troubleshooting

Tests hanging or taking too long

Linter too strict

Type checker too strict

Customization Notes

Use it

Manifest

Reviews

About

Quality Check and Fix

Objective

Context

Workflow

Step 1: Run Tests

Step 2: Run Linting (Optional)

Step 3: Run Type Checking (Optional)

Step 4: Run Makefile Target (If Available)

Fixing Failures

Test Failures

Lint Failures

Type Errors

Constraints

From Repository Instructions

Best Practices

Example Sessions

Example 1: Tests pass, lint has minor issues

Example 2: Test failures due to missing dependency

Example 3: Complex test failure requiring code fix

Output Format

Success Report

Failure Report

Makefile Integration

If make check exists:

Makefile check target example:

Troubleshooting

Tests hanging or taking too long

Linter too strict

Type checker too strict

Customization Notes

Use it

Manifest

Reviews

If `make check` exists: