deadbits/vigil-llm
⚡ Vigil ⚡ Detect prompt injections, jailbreaks, and other potentially risky Large Language Model (LLM) inputs
This project helps security and operations teams protect their AI applications by detecting harmful or malicious inputs to Large Language Models (LLMs). It takes user prompts and LLM responses as input and identifies potential prompt injections, jailbreaks, or other security threats, providing an assessment of the risk. Anyone responsible for the security and integrity of applications built on LLMs, especially those in cybersecurity or AI operations roles, would use this.
464 stars. No commits in the last 6 months.
Use this if you are developing or managing an application that uses Large Language Models and need to protect it from common attack vectors like prompt injections and jailbreaks.
Not ideal if you need an enterprise-grade, fully supported AI firewall solution, as this project is currently in an experimental alpha state.
Stars
464
Forks
53
Language
Python
License
Apache-2.0
Category
Last pushed
Jan 31, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/deadbits/vigil-llm"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ethz-spylab/agentdojo
A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.
guardrails-ai/guardrails
Adding guardrails to large language models.
JasonLovesDoggo/caddy-defender
Caddy module to block or manipulate requests originating from AIs or cloud services trying to...
inkdust2021/VibeGuard
Uses just 1% memory while protecting 99% of your personal privacy.
Heiberg-Industries/designbrief
Design guardrails, not templates. A library of UI design direction files that give LLMs (and...