Copy this prompt into our developer Console to try it for yourself!

Content
SystemYour task is to create Python functions based on the provided natural language requests. The requests will describe the desired functionality of the function, including the input parameters and expected return value. Implement the functions according to the given specifications, ensuring that they handle edge cases, perform necessary validations, and follow best practices for Python programming. Please include appropriate comments in the code to explain the logic and assist other developers in understanding the implementation.
UserI want a function that can solve a Sudoku puzzle. The function should take a 9x9 Sudoku grid as input, where empty cells are represented by the value 0. The function should solve the puzzle using a backtracking algorithm and return the solved grid. If the puzzle is unsolvable, it should return None. The function should also validate the input grid to ensure it is a valid Sudoku puzzle.

​
Example Output

Python
def solve_sudoku(grid):
   def is_valid(row, col, num):
       # Check if the number is valid in the current row and column
       for i in range(9):
           if grid[row][i] == num or grid[i][col] == num:
               return False

       # Check if the number is valid in the current 3x3 subgrid
       start_row, start_col = (row // 3) * 3, (col // 3) * 3
       for i in range(3):
           for j in range(3):
               if grid[start_row + i][start_col + j] == num:
                   return False

       return True

   def backtrack(row, col):
       # If we have filled all cells, the puzzle is solved
       if row == 9:
           return True

       # Move to the next row if we have reached the end of the current row
       if col == 9:
           return backtrack(row + 1, 0)

       # If the current cell is already filled, move to the next cell
       if grid[row][col] != 0:
           return backtrack(row, col + 1)

       # Try filling the current cell with numbers from 1 to 9
       for num in range(1, 10):
           if is_valid(row, col, num):
               grid[row][col] = num
               if backtrack(row, col + 1):
                   return True
               grid[row][col] = 0

       # If no valid number is found, backtrack
       return False

   # Validate the input grid
   if not all(len(row) == 9 and all(0 <= cell <= 9 for cell in row) for row in grid):
       return None

   # Start the backtracking algorithm
   if backtrack(0, 0):
       return grid
   else:
       return None

​
API Request

import anthropic

client = anthropic.Anthropic(  # defaults to os.environ.get("ANTHROPIC_API_KEY")
    api_key="my_api_key",
)
message = client.messages.create(
    model="claude-opus-4-20250514",
    max_tokens=1000,
    temperature=0,
    system="Your task is to create Python functions based on the provided natural language requests. The requests will describe the desired functionality of the function, including the input parameters and expected return value. Implement the functions according to the given specifications, ensuring that they handle edge cases, perform necessary validations, and follow best practices for Python programming. Please include appropriate comments in the code to explain the logic and assist other developers in understanding the implementation.",
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "I want a function that can solve a Sudoku puzzle. The function should take a 9x9 Sudoku grid as input, where empty cells are represented by the value 0. The function should solve the puzzle using a backtracking algorithm and return the solved grid. If the puzzle is unsolvable, it should return None. The function should also validate the input grid to ensure it is a valid Sudoku puzzle.",
                }
            ],
        }
    ],
)
print(message.content)