CLAUDE FOR CODING

Claude's Coding Capabilities:

The Claude Coding Workflow:

1. PLAN
   Claude: "Design a REST API for user management"
   Output: Endpoint list, data models, authentication strategy

2. IMPLEMENT
   Claude: "Generate FastAPI endpoints with JWT auth"
   Output: Working code with error handling

3. TEST
   Claude: "Write pytest unit tests for these endpoints"
   Output: Test suite with edge cases

4. REVIEW
   Claude: "Review this code for security vulnerabilities"
   Output: Security audit + fixes

5. REFACTOR
   Claude: "Refactor to use async/await and connection pooling"
   Output: Optimized production-ready code

6. DEPLOY
   Claude: "Generate Dockerfile and docker-compose.yml"
   Output: Containerization configs

7. DOCUMENT
   Claude: "Generate OpenAPI spec and README"
   Output: Complete documentation

Claude vs Human vs GPT-4 for Coding:

Pattern 1: Specification → Implementation

FUNCTIONAL REQUIREMENTS:
- [Requirement 1]
- [Requirement 2]
- [Requirement 3]

CONSTRAINTS:
- [Constraint 1]
- [Constraint 2]

OUTPUT:
- Complete, working code
- Comments explaining complex logic
- Error handling for edge cases
- Input validation

Real Example: Rate-Limited API Client

User: Implement a rate-limited API client in Python that calls the Claude API.

REQUIREMENTS:
- Max 50 requests per minute (token bucket algorithm)
- Automatic retry on 429 (rate limit) with exponential backoff
- Async support (asyncio)
- Logging for debugging
- Environment variable for API key

CONSTRAINTS:
- Use aiohttp, no requests library
- Include type hints
- Maximum 150 lines of code

OUTPUT: Complete working class.

Claude's Response (abbreviated - ~150 lines of working code):

import asyncio
import logging
import os
import time
from typing import Optional, Dict, Any
import aiohttp

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

class TokenBucket:
    """Token bucket rate limiter implementation"""

    def __init__(self, rate: int, capacity: int):
        self.rate = rate  # tokens per second
        self.capacity = capacity
        self.tokens = capacity
        self.last_refill = time.monotonic()

    async def acquire(self) -> bool:
        now = time.monotonic()
        elapsed = now - self.last_refill
        self.tokens = min(self.capacity, self.tokens + elapsed * self.rate)
        self.last_refill = now

        if self.tokens >= 1:
            self.tokens -= 1
            return True
        else:
            wait_time = (1 - self.tokens) / self.rate
            await asyncio.sleep(wait_time)
            return await self.acquire()

class RateLimitedClaudeClient:
    """Async Claude API client with rate limiting"""

    def __init__(self, api_key: Optional[str] = None,
                 max_requests_per_minute: int = 50,
                 max_retries: int = 5):

        self.api_key = api_key or os.environ.get("ANTHROPIC_API_KEY")
        if not self.api_key:
            raise ValueError("ANTHROPIC_API_KEY not found")

        self.base_url = "<https://api.anthropic.com/v1/messages>"
        self.rate_limiter = TokenBucket(
            rate=max_requests_per_minute / 60.0,
            capacity=max_requests_per_minute
        )
        self.max_retries = max_retries

    async def generate(self, prompt: str, model: str = "claude-3-sonnet-20241022",
                      max_tokens: int = 1000, temperature: float = 0.7) -> str:

        await self.rate_limiter.acquire()

        # ... implementation with retries and error handling

The Systematic Debugging Prompt:

I'm getting this error in my {language} code:

ERROR:
[Full error message with stack trace]

CODE:
```{language}
[Relevant code section, 20-50 lines]

CONTEXT:

• Framework: {framework} version {version}
• Dependencies: {list relevant packages}
• What I've tried: [1, 2, 3]
• When it works: {e.g., works locally, fails in production}

Please help me debug by:

1. EXPLAINING the root cause
2. PROVIDING the fix
3. EXPLAINING why the fix works
4. SUGGESTING how to prevent similar issues

**Real Debugging Example:**

User: ERROR in Node.js/Express:

ERROR:

TypeError: Cannot read property 'id' of undefined

at app.post (/app/routes/users.js:45:28)

at Layer.handle [as handle_request] (/app/node_modules/express/lib/router/layer.js:95:5)

CODE:

javascript

app.post('/api/users/:id', (req, res) => {
    const userId = req.params.id;
    const userData = req.body;

    // Validate user exists
    if (!req.user.id == userId) {  // LINE 45
        return res.status(403).json({ error: 'Unauthorized'});
    }

    const updatedUser = database.updateUser(userId, userData);
    res.json(updatedUser);
});

CONTEXT:

• Express 4.18
• Using JWT authentication middleware (attaches user to req.user)
• Error occurs on some requests, not all
• Works when user is properly authenticated

Claude:

<root_cause>

The error occurs because req.user is undefined on some requests.

Line 45: if (!req.user.id == userId) tries to access id property of req.user,

but when authentication middleware fails, req.user is never set and remains undefined.

The condition logic is also flawed: !req.user.id evaluates to !undefined → true,

then true === userId (string) → false, so the error appears BEFORE the condition

can fail safely.

</root_cause>

<fix> ```javascript app.post('/api/users/:id', (req, res) => { const userId = req.params.id; const userData = req.body;

// First check if user exists

if (!req.user) {

return res.status(401).json({ error: 'Authentication required' });

}

// Then check authorization

if (req.user.id !== userId) {

return res.status(403).json({ error: 'Unauthorized access to this user' });

}

const updatedUser = database.updateUser(userId, userData);

res.json(updatedUser);

});

</fix>

<why_it_works>
1. The req.user check fails early if middleware didn't set the user
2. Changed comparison to !== (removed the ! which was causing truthy issues)
3. Added specific error messages for each failure case
</why_it_works>

<prevention>
To prevent this class of error:
1. Ensure authentication middleware ALWAYS sets req.user (or null)
2. Use optional chaining: req.user?.id
3. Add validation middleware that guarantees authentication state
</prevention>

Test Generation Prompt Template:

Generate pytest unit tests for this {function/class}.

CODE UNDER TEST:
```{language}
[Code to test]

REQUIREMENTS:

• Coverage: 80-100%
• Test framework: pytest
• Include fixtures for setup
• Mock external dependencies (database, API, filesystem)
• Test these categories:
  • Happy path (normal inputs)
  • Edge cases (empty, max, min, null)
  • Error cases (invalid input, exceptions)