@presidio-dev/hai-guardrails

A set of guards for LLM Apps

v1.11.1 URL: https://unpkg.com/@presidio-dev/hai-guardrails@1.11.1/dist/index.cjs

Open Browse Files

presidio guardrails llm hai redaction security defence governance guards human-ai prompt-injection llm-guardrails halucination

spotlighting-datamarking

This is a package to implement data marking functionality to make indirect prompt injections difficult, based on the research done by Microsoft

v1.0.0-alpha URL: https://unpkg.com/spotlighting-datamarking@1.0.0-alpha/index.js

Open Browse Files

security prompt-injection llm ai-security spotlighting data-marking prompt-protection microsoft-research

aidefence

AI Defence - Production-ready adversarial defense system for AI applications

v2.1.1 URL: https://unpkg.com/aidefence@2.1.1/dist/index.js

Open Browse Files

llm-guard

A TypeScript library for validating and securing LLM prompts

v0.1.8 URL: https://unpkg.com/llm-guard@0.1.8/dist/index.js

Open Browse Files

llm security validation prompt jailbreak pii toxicity profanity prompt-injection relevance

aidefense

AI Defense - Wrapper for aidefence (American spelling). Production-ready adversarial defense system for AI applications with real-time threat detection, behavioral analysis, and formal verification.

v2.1.1 URL: https://unpkg.com/aidefense@2.1.1/index.js

Open Browse Files

ai security llm prompt-injection defense wasm detection analysis verification neuro-symbolic multimodal adversarial-defense jailbreak-detection agentdb midstream aidefence wrapper