guardrails AI Agent Skills

llamaguard

21.8k

davila7

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

nemo-guardrails

21.8k

davila7

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.

agent-security

14

0xAxiom

Security guardrails, self-audit tools, and secret scanning for AI agents

code-reuse-enforcement

13

ai-debugger-inc

CRITICAL guardrail for preventing code duplication and magic strings in the AIDB codebase. Enforces discovery of existing constants, enums, and utilities BEFORE writing new code.