Goal

Make the AI produce clear, efficient, well-structured, and maintainable Python code with good reasoning and explanations.

---

1. Understand the Problem

• Begin by summarizing the task in simple terms.
• Outline the approach before writing code.
• Break down complex problems into smaller manageable parts.

2. Code Style and Best Practices

• Follow PEP8 naming and formatting conventions.
• Use descriptive variable and function names.
• Avoid unnecessary complexity; prefer clarity over cleverness.
• Keep functions short and focused on a single task.

3. Structure and Modularity

• Use functions and classes to organize code logically.
• Separate concerns: input handling, processing, output.
• Use comments sparingly but meaningfully to explain non-obvious parts.

4. Reasoning in the Code Generation

• Before coding, write a brief plan or algorithm outline.
• Mention key data structures or libraries to be used.
• Anticipate edge cases or error conditions and address them.

5. Explanations and Documentation

• Provide concise explanations for key steps or algorithms.
• Use docstrings for functions and classes explaining purpose and parameters.
• When returning results, explain what the output represents.

6. Testing and Validation

• Include basic tests or usage examples when appropriate.
• Validate inputs if necessary to avoid runtime errors.

7. Efficiency and Optimization

• Write clear code first; optimize only if needed.
• Use built-in functions and libraries when possible.

---

Example Workflow

1. Describe problem: "Implement a function to find the nth Fibonacci number using iteration."
2. Outline approach: "Use a loop to calculate Fibonacci numbers up to n, storing only last two values for efficiency."
3. Write code with docstrings and variable names.
4. Explain key steps briefly after the code.
5. Include a simple test call at the end.