# Guardrails — Validate & Secure LLM Outputs

> Guardrails is a Python framework for validating LLM inputs/outputs to detect risks and generate structured data. 6.6K+ GitHub stars. Pre-built validators, Pydantic models. Apache 2.0.

## Install

Save as a script file and run:

## Quick Use

```bash
# Install
pip install guardrails-ai

# Example: validate LLM output as structured data
python -c "
import guardrails as gd
from pydantic import BaseModel

class Pet(BaseModel):
    name: str
    species: str
    age: int

guard = gd.Guard.from_pydantic(Pet)
result = guard(
    model='gpt-4o-mini',
    messages=[{'role': 'user', 'content': 'Tell me about a pet.'}]
)
print(result.validated_output)  # {'name': 'Buddy', 'species': 'Dog', 'age': 3}
"
```

---

## Intro

Guardrails is a Python framework for building reliable AI applications by validating LLM inputs and outputs to detect and mitigate risks, and generate structured data. With 6,600+ GitHub stars and Apache 2.0 license, it provides pre-built validators through Guardrails Hub covering common risk categories (PII, toxicity, hallucination, SQL injection), input/output guards that intercept LLM interactions, structured data generation using Pydantic models, standalone server deployment via REST API, and support for both proprietary and open-source LLMs.

**Best for**: Teams building production AI apps who need output validation, safety guardrails, and structured outputs
**Works with**: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf
**LLMs**: OpenAI, Anthropic, Cohere, HuggingFace, and any LLM

---

## Key Features

- **Pre-built validators**: PII detection, toxicity, hallucination, SQL injection via Hub
- **Input/output guards**: Intercept and validate every LLM interaction
- **Structured outputs**: Generate typed data using Pydantic models
- **REST API server**: Deploy guards as a standalone service
- **Custom validators**: Create domain-specific validation rules
- **Any LLM**: Works with proprietary and open-source models

---

### FAQ

**Q: What is Guardrails?**
A: Guardrails is a Python framework with 6.6K+ stars for validating LLM inputs/outputs. Pre-built validators for PII, toxicity, hallucination. Structured outputs via Pydantic. Apache 2.0.

**Q: How do I install Guardrails?**
A: `pip install guardrails-ai`. Use `Guard.from_pydantic(Model)` to create validated LLM calls.

---

## Source & Thanks

> Created by [Guardrails AI](https://github.com/guardrails-ai). Licensed under Apache 2.0.
> [guardrails-ai/guardrails](https://github.com/guardrails-ai/guardrails) — 6,600+ GitHub stars

---
Source: https://tokrepo.com/en/workflows/f10382e2-4fa6-491f-8110-84f8397af129
Author: Script Depot