guardrails-lite-client

A bare minimum deployment showing how to use guardrails client side (in an application). This deployment contains a simple OpenAI compliant chat completions endpoint. The completions are then validated by a Guard using the ToxicLanguage Validator.

Run the server locally with docker

Linux and MacOS

1. Clone this repository
2. Make sure there is a valid Guardrails AI API Key in the GUARDRAILS_TOKEN environment variable. ./buildscripts/build/.sh uses this for performing hub installs during the Docker build.
3. make build
4. make start

Once the server is up and running, you can check out the Swagger docs at http://localhost:8000/docs

Productionizing

We include a Dockerfile that shows the basic steps of containerizing this server.

We also include two bash scripts in the buildscripts directory that shows the basics of building the image and running it.