Defend

Getting started

Install pydefend, configure the server, and learn the basic request flow for input and output guarding.

View on GitHub

This section walks you from installation to a working local API. You run Defend as a small FastAPI service (defend serve), configure it with defend.config.yaml, and call /v1/guard/input before your LLM and /v1/guard/output after.

Installation

Install from PyPI with pip, optional Docker image, and environment setup.

Quick start

Minimal config, defend serve, and curl (or Python) against the guard endpoints.

If you are new to sessions and actions, read Sessions next.

FAQ

Introduction

AI security guardrails for LLM applications with input and output guarding, sessions, and configurable modules.

Installation

Install the pydefend package from PyPI and prepare your environment for running the Defend API locally or in Docker.

On this page

FAQ