How to Validate LLM Output in Python

import json
from typing import Any, Dict, List, Optional


class DataValidator:
    """Simple helper for validating LLM-generated or user data."""

    def __init__(self, required_fields: List[str], schema: Optional[Dict[str, str]] = None):
        self.required_fields = required_fields
        self.schema = schema or {}

    def validate(self, data: Dict[str, Any]) -> Dict[str, Any]:
        """Validate data and return normalized result with status."""
        errors = []

        for field in self.required_fields:
            if field not in data or data[field] in (None, ""):
                errors.append(f"Missing required field: {field}")

        for field, expected_type in self.schema.items():
            if field in data and data[field] is not None:
                if not isinstance(data[field], eval(expected_type)):
                    errors.append(
                        f"Field '{field}' should be {expected_type}, got {type(data[field]).__name__}"
                    )

        return {
            "is_valid": len(errors) == 0,
            "errors": errors,
            "data": self._clean(data),
        }

    def _clean(self, data: Dict[str, Any]) -> Dict[str, Any]:
        return {k: v for k, v in data.items() if v is not None}


if __name__ == "__main__":
    # Example: validating LLM response for a user profile
    validator = DataValidator(
        required_fields=["name", "email"],
        schema={"age": "int", "active": "bool"},
    )

    raw_llm_output = json.loads('{"name": "Alice", "email": "alice@example.com", "age": "30"}')
    result = validator.validate(raw_llm_output)

    print(json.dumps(result, indent=2))