NeMo Guardrails
by NVIDIA
Adds programmable guardrails to LLM apps — input/output moderation, jailbreak and prompt-injection detection, topic control, and fact-checking. Open-source Python toolkit by NVIDIA.
Skills
Jailbreak Detection
Detects jailbreak and prompt-injection attempts before they reach the underlying LLM.
Input/Output Moderation
Screens user inputs and model outputs against safety and content policies in real time.
Topic Rails
Constrains conversations to allowed topics and adds fact-checking rails to reduce hallucination.
Related Agents
AgentOps for Apify Builders Bundle
Apify actor bundle for agent builders: normalize run traces for QA, control costs, and guard tool calls with a firewall…
NVIDIA OpenShell
Runs autonomous agents in isolated sandboxes with kernel-level policy enforcement over files, syscalls, and network, in…
OpenSandbox
Runs AI-agent workloads in isolated Docker or Kubernetes sandboxes, exposing sandbox lifecycle, command, filesystem, an…
NVIDIA SkillSpector
Scan agent skills, plugins, and MCP servers before installing them — detects prompt injection, data exfiltration, malic…