---
name: implementation
description: Implement features with code, tests, and documentation. Use when building
  features from approved designs following TDD and project coding standards.
allowed-tools: Read, Write, Edit, Bash, Grep, Glob
---

# Feature Implementation Skill

## Purpose

This skill provides systematic guidance for implementing features with high-quality code, comprehensive tests, and proper documentation, following project standards and best practices.

## When to Use

- After design phase is complete and approved
- Need to implement code for a feature
- Writing unit and integration tests
- Creating technical documentation
- Following TDD (Test-Driven Development) workflow

## Implementation Workflow

### 1. Setup and Preparation

**Review Design Document:**
- Read architecture design from previous phase
- Understand component structure
- Review API contracts and data models
- Note security and performance requirements

**Setup Development Environment:**
```bash
# Activate virtual environment
source venv/bin/activate  # or: uv venv && source .venv/bin/activate

# Install dependencies
pip install -r requirements.txt
# or: uv pip install -r requirements.txt

# Install dev dependencies
pip install -e ".[dev]"
```

**Create Feature Branch:**
```bash
git checkout -b feature/feature-name
```

**Deliverable:** Development environment ready

---

### 2. Test-Driven Development (TDD)

**TDD Cycle:** Red → Green → Refactor

**Step 1: Write Failing Test (Red)**
```python
# tests/test_feature.py
import pytest
from feature import process_data

def test_process_data_success():
    """Test successful data processing."""
    # Arrange
    input_data = {"name": "test", "value": 123}

    # Act
    result = process_data(input_data)

    # Assert
    assert result.name == "test"
    assert result.value == 123
```

**Step 2: Write Minimal Code (Green)**
```python
# src/tools/feature/core.py
def process_data(input_data: dict):
    """Process input data."""
    # Minimal implementation to pass test
    return type('Result', (), input_data)()
```

**Step 3: Refactor (Refactor)**
```python
# src/tools/feature/core.py
from .models import InputModel, ResultModel

def process_data(input_data: dict) -> ResultModel:
    """
    Process input data and return result.

    Args:
        input_data: Input data dictionary

    Returns:
        ResultModel with processed data

    Raises:
        ValidationError: If input is invalid
    """
    # Proper implementation with validation
    validated = InputModel(**input_data)
    return ResultModel(
        name=validated.name,
        value=validated.value
    )
```

**Repeat:** Write next test, implement, refactor

**Deliverable:** Tested, working code

---

### 3. Code Implementation

**Follow Project Structure:**
```
src/tools/feature_name/
├── __init__.py           # Public exports
├── models.py             # Pydantic models (data)
├── interfaces.py         # Abstract interfaces
├── core.py               # Core business logic
├── repository.py         # Data access layer
├── validators.py         # Input validation
├── utils.py              # Helper functions
├── config.py             # Configuration
├── exceptions.py         # Custom exceptions
└── main.py               # CLI entry point (if applicable)
```

**Coding Standards:**
Refer to `code-style-guide.md` for:
- PEP 8 style guide
- Type hints for all functions
- Google-style docstrings
- 500-line file limit
- Single responsibility principle

**Example Implementation:**
```python
# src/tools/feature/models.py
from pydantic import BaseModel, Field
from typing import Optional
from datetime import datetime

class FeatureInput(BaseModel):
    """Input model for feature."""

    name: str = Field(..., min_length=1, max_length=100)
    value: int = Field(..., ge=0)

    class Config:
        validate_assignment = True


class FeatureOutput(BaseModel):
    """Output model for feature."""

    id: Optional[int] = None
    name: str
    value: int
    created_at: datetime = Field(default_factory=datetime.utcnow)


# src/tools/feature/core.py
from .models import FeatureInput, FeatureOutput
from .repository import FeatureRepository
from .validators import FeatureValidator

class FeatureService:
    """Feature service with business logic."""

    def __init__(
        self,
        repository: FeatureRepository,
        validator: FeatureValidator
    ):
        """
        Initialize service with dependencies.

        Args:
            repository: Repository for data access
            validator: Validator for input validation
        """
        self.repository = repository
        self.validator = validator

    def create(self, input_data: FeatureInput) -> FeatureOutput:
        """
        Create new feature resource.

        Args:
            input_data: Validated input data

        Returns:
            FeatureOutput with created resource

        Raises:
            ValidationError: If validation fails
            RepositoryError: If save fails
        """
        # Validate
        self.validator.validate_create(input_data)

        # Create
        output = FeatureOutput(
            name=input_data.name,
            value=input_data.value
        )

        # Persist
        saved = self.repository.save(output)

        return saved
```

**Deliverable:** Implemented core functionality

---

### 4. Testing Implementation

**Testing Checklist:**
Refer to `testing-checklist.md` for comprehensive coverage

**Unit Tests (80%+ Coverage):**
```python
# tests/test_core.py
import pytest
from unittest.mock import Mock, MagicMock
from feature.core import FeatureService
from feature.models import FeatureInput, FeatureOutput

@pytest.fixture
def mock_repository():
    """Mock repository for testing."""
    repo = Mock()
    repo.save.return_value = FeatureOutput(
        id=1,
        name="test",
        value=123
    )
    return repo

@pytest.fixture
def mock_validator():
    """Mock validator for testing."""
    validator = Mock()
    validator.validate_create.return_value = None
    return validator

@pytest.fixture
def service(mock_repository, mock_validator):
    """Service fixture with mocked dependencies."""
    return FeatureService(
        repository=mock_repository,
        validator=mock_validator
    )

def test_create_success(service, mock_repository):
    """Test successful creation."""
    # Arrange
    input_data = FeatureInput(name="test", value=123)

    # Act
    result = service.create(input_data)

    # Assert
    assert result.name == "test"
    assert result.value == 123
    mock_repository.save.assert_called_once()

def test_create_validation_error(service, mock_validator):
    """Test validation error handling."""
    # Arrange
    input_data = FeatureInput(name="test", value=123)
    mock_validator.validate_create.side_effect = ValidationError("Invalid")

    # Act & Assert
    with pytest.raises(ValidationError):
        service.create(input_data)
```

**Integration Tests:**
```python
# tests/integration/test_feature_integration.py
import pytest
from pathlib import Path
from feature import FeatureService, FileSystemRepository

@pytest.fixture
def temp_data_dir(tmp_path):
    """Temporary directory for test data."""
    return tmp_path / "data"

def test_create_and_retrieve(temp_data_dir):
    """Test end-to-end create and retrieve."""
    # Arrange
    repo = FileSystemRepository(temp_data_dir)
    service = FeatureService(repo)

    # Act: Create
    created = service.create(FeatureInput(name="test", value=123))

    # Act: Retrieve
    retrieved = service.get(created.id)

    # Assert
    assert retrieved.name == "test"
    assert retrieved.value == 123
```

**Run Tests:**
```bash
# Run all tests with coverage
pytest --cov=src --cov-report=html --cov-report=term

# Run specific test file
pytest tests/test_core.py -v

# Run with markers
pytest -m "not slow" -v
```

**Deliverable:** Comprehensive test suite (80%+ coverage)

---

### 5. Code Quality Checks

**Run Formatters and Linters:**
```bash
# Format code with Black
black src/ tests/

# Type check with mypy
mypy src/

# Lint with flake8 (if configured)
flake8 src/ tests/

# Run all checks
make lint  # If Makefile configured
```

**Pre-commit Hooks (If Configured):**
```bash
# Run pre-commit checks
pre-commit run --all-files
```

**Code Review Checklist:**
- [ ] All functions have type hints
- [ ] All functions have docstrings
- [ ] No files exceed 500 lines
- [ ] Tests achieve 80%+ coverage
- [ ] No lint errors or warnings
- [ ] Error handling implemented
- [ ] Logging added where appropriate
- [ ] Security best practices followed

**Deliverable:** Quality-checked code

---

### 6. Documentation

**Code Documentation:**
- Docstrings for all public functions/classes
- Inline comments for complex logic
- Type hints for clarity

**Technical Documentation:**
```markdown
# Feature Implementation

## Overview
[What was implemented]

## Architecture
[Actual structure (may differ from design)]

## Usage Examples
```python
from feature import FeatureService

service = FeatureService()
result = service.create(name="example")
```

## Configuration
Required environment variables:
- `FEATURE_API_KEY`: API key for service
- `FEATURE_TIMEOUT`: Timeout in seconds (default: 30)

## Testing
```bash
pytest tests/test_feature.py
```

## Known Issues
- [Issue 1]: [Workaround]

## Future Enhancements
- [Enhancement 1]
```

**User Documentation (If Applicable):**
- Usage guide in `docs/guides/`
- CLI help text
- Example configurations

**Deliverable:** Complete documentation

---

### 7. Integration and Verification

**Verify Against Requirements:**
- [ ] All acceptance criteria met
- [ ] Security checklist items addressed
- [ ] Performance requirements met
- [ ] Edge cases handled
- [ ] Error scenarios tested

**Manual Testing:**
```bash
# Test CLI (if applicable)
python -m src.tools.feature.main create --name test

# Test with real data
python -m src.tools.feature.main --input sample.json

# Test error cases
python -m src.tools.feature.main --invalid-input
```

**Integration with Existing Code:**
- [ ] Imports work correctly
- [ ] No circular dependencies
- [ ] Backward compatibility maintained (if applicable)
- [ ] No breaking changes to public APIs

**Deliverable:** Verified, working feature

---

## Code Style Guidelines

### Python Style (PEP 8)

**Imports:**
```python
# Standard library
import os
import sys
from pathlib import Path

# Third-party
import click
from pydantic import BaseModel

# Local
from .models import FeatureModel
from .exceptions import FeatureError
```

**Naming:**
```python
# Classes: PascalCase
class FeatureService:
    pass

# Functions/methods: snake_case
def process_data():
    pass

# Constants: UPPER_SNAKE_CASE
MAX_RETRIES = 3

# Private: leading underscore
def _internal_helper():
    pass
```

**Type Hints:**
```python
from typing import Optional, List, Dict, Union

def function(
    required: str,
    optional: Optional[int] = None,
    items: List[str] = None
) -> Dict[str, Any]:
    pass
```

**Docstrings (Google Style):**
```python
def function(param1: str, param2: int) -> bool:
    """
    Short description.

    Longer description if needed.

    Args:
        param1: Description of param1
        param2: Description of param2

    Returns:
        Description of return value

    Raises:
        ValueError: When this happens
    """
    pass
```

---

## Testing Best Practices

### Pytest Conventions

**Test File Naming:**
- `test_*.py` or `*_test.py`
- Mirror source structure: `src/core.py` → `tests/test_core.py`

**Test Function Naming:**
```python
def test_function_name_condition_expected_result():
    """Test description."""
    pass

# Examples:
def test_create_feature_valid_input_returns_feature():
    pass

def test_validate_input_missing_name_raises_error():
    pass
```

**Test Structure (Arrange-Act-Assert):**
```python
def test_example():
    """Test example."""
    # Arrange: Setup test data and mocks
    input_data = {"name": "test"}
    mock_service = Mock()

    # Act: Execute the code being tested
    result = function_under_test(input_data, mock_service)

    # Assert: Verify expected outcomes
    assert result == expected
    mock_service.method.assert_called_once()
```

**Fixtures:**
```python
# tests/conftest.py (shared fixtures)
import pytest

@pytest.fixture
def sample_data():
    """Sample data for tests."""
    return {"name": "test", "value": 123}

@pytest.fixture
def temp_directory(tmp_path):
    """Temporary directory for test files."""
    test_dir = tmp_path / "test_data"
    test_dir.mkdir()
    yield test_dir
    # Cleanup happens automatically
```

**Parametrize for Multiple Cases:**
```python
@pytest.mark.parametrize("input_value,expected", [
    ("valid", True),
    ("invalid", False),
    ("", False),
])
def test_validation(input_value, expected):
    """Test validation with multiple inputs."""
    result = validate(input_value)
    assert result == expected
```

---

## Common Patterns

### Error Handling Pattern

```python
from typing import Optional
import logging

logger = logging.getLogger(__name__)

def process_data(data: dict) -> Result:
    """Process data with proper error handling."""
    try:
        # Validate
        validated = validate_data(data)

        # Process
        result = perform_processing(validated)

        return result

    except ValidationError as e:
        logger.warning(f"Validation failed: {e}")
        raise

    except ProcessingError as e:
        logger.error(f"Processing failed: {e}", exc_info=True)
        raise

    except Exception as e:
        logger.exception(f"Unexpected error: {e}")
        raise ProcessingError("Unexpected error occurred") from e
```

### Dependency Injection Pattern

```python
from abc import ABC, abstractmethod

# Interface
class Repository(ABC):
    @abstractmethod
    def save(self, data) -> None:
        pass

# Implementation
class FileRepository(Repository):
    def save(self, data) -> None:
        # File-based implementation
        pass

# Service with dependency injection
class Service:
    def __init__(self, repository: Repository):
        self.repository = repository  # Injected dependency

    def create(self, data):
        # Use injected repository
        self.repository.save(data)

# Usage
repo = FileRepository()
service = Service(repository=repo)  # Inject dependency
```

### Configuration Pattern

```python
from pydantic_settings import BaseSettings

class Config(BaseSettings):
    """Application configuration."""

    api_key: str
    timeout: int = 30
    debug: bool = False

    class Config:
        env_prefix = "FEATURE_"
        env_file = ".env"

# Usage
config = Config()  # Loads from environment/file
service = Service(api_key=config.api_key, timeout=config.timeout)
```

---

## Supporting Resources

- **code-style-guide.md**: Detailed Python style guidelines
- **testing-checklist.md**: Comprehensive testing requirements
- **scripts/generate_tests.py**: Test scaffolding automation

---

## Integration with Feature Implementation Flow

**Input:** Approved architecture design
**Process:** TDD implementation with quality checks
**Output:** Tested, documented code
**Next Step:** Validation skill for quality assurance

---

## Implementation Checklist

Before marking feature complete:
- [ ] All code implemented per design
- [ ] Unit tests written (80%+ coverage)
- [ ] Integration tests written
- [ ] All tests passing
- [ ] Code formatted (Black)
- [ ] Type checking passing (mypy)
- [ ] No lint errors
- [ ] Docstrings complete
- [ ] Technical documentation written
- [ ] User documentation written (if applicable)
- [ ] Manual testing completed
- [ ] Security considerations addressed
- [ ] Performance requirements met
- [ ] Code reviewed (if applicable)
- [ ] Ready for validation phase