AI/TLDR

Protect AI · 2024-08-10 · notable

LLM Guard — Input/Output Security Toolkit

Self-hosted security toolkit for LLM interactions. Scans inputs for prompt injection, jailbreaks, and toxic content. Scans outputs for PII, secrets, and harmful content.

LLM Guard security toolkit repository

Scan LLM inputs and outputs for security risks — self-hosted, no data leaves your infra.

Key specs

GitHub stars1.8k+
Scanners20+

What is it?

LLM Guard is a self-hosted security layer for LLM apps. It runs before and after every LLM call, scanning for prompt injections, jailbreaks, PII leakage, secrets, and harmful content.

How does it work?

Deploy it as a Python library or API service. Configure input scanners (prompt injection, toxicity, language) and output scanners (PII, secrets, relevance). It blocks or sanitizes risky content.

Why does it matter?

Unlike cloud-based solutions, LLM Guard runs entirely on your infrastructure. Sensitive data never leaves your network.

Who is it for?

Security-conscious teams who need on-prem LLM guardrails.

Try it

pip install llm-guard

Sources · 2 outlets

Tags

  • llm-guard
  • prompt-injection
  • pii-detection
  • self-hosted
  • open-source

← All releases · Learn AI