A framework that finds and masks personal data in text, images and tables. It detects the data, then redacts or replaces it so it can be safely sent to a model.
All repositories
Securitymeta-llama
PurpleLlama
Meta's generative AI safety project: the Llama Guard and Prompt Guard filter models, the Code Shield scanner and the CyberSec Eval test suites. It pairs defense with attack-based testing.
Who it is for
For people who want to filter model inputs and outputs and evaluate cybersecurity risk.
How to start
- Read System-Level Safeguards and pick a component: Llama Guard, Prompt Guard or Code Shield
- Use the Llama-recipes repo for setup of the safeguard models (linked under Getting Started)
- Explore the CyberSec Eval suites to measure your model's risks
Steps are taken from the README. Check the current version in the repository before running them.
Stars over the last 30 days
+45Sep 6 — Oct 6
Author's description
Set of tools to assess and improve LLM security.
More in «Security»
A vulnerability scanner for LLMs: it runs a model through a set of probes for prompt injection, data leakage, toxicity and other failures. Works as a command-line tool.
A Python framework that checks LLM inputs and outputs with ready validators from Guardrails Hub and helps produce structured answers. It catches risks before a reply reaches the user.
NVIDIA's toolkit for programmable rails in LLM-based conversational systems. Rules live in config and control what the bot talks about and does.
Microsoft's framework for proactively finding risks in generative AI systems. It helps automate red teaming against models and applications.
A security scanner for AI agents, MCP servers and skills. It checks configs and SKILL.md files for risky spots.