AuditaAI
Back to vendors

AI security vendor

NVIDIA Corporation

Commercial parent organization

Official website

4

Offerings

4

Reviewed claims

5

Capabilities

Domains

Agent SecurityModel and ValidationRuntime Protection

Actions

PREVENTPOSTUREAUDIT

Offerings

Open an offering for reviewed behavior and evidence

NVIDIA Corporation · project

NVIDIA NeMo Guardrails

1 claims+

Open-source toolkit for programmable dialog, topical, and safety controls in LLM applications.

Official source
Prompt and Response Guardrails
PREVENT

NeMo Guardrails applies programmable dialog, topical, and safety rails to inspect and constrain input prompts and model responses in LLM applications.

Source

NVIDIA Corporation · project

garak

1 claims+

Open-source vulnerability scanner that probes language models for security weaknesses.

Official source
Model Security Testing
POSTURE

garak probes language-model endpoints across attack classes including jailbreaks, prompt injection, and data leakage to identify model security weaknesses.

Source

NVIDIA Corporation · project

NVIDIA NemoClaw

1 claims+

Open-source reference stack that packages supported autonomous agents with hardened blueprints, policy-controlled sandboxing, routed inference, and lifecycle tooling.

Official source
Agent Execution IsolationAgent-Human Oversight
PREVENT

NemoClaw deploys supported autonomous agents from a digest-verified hardened blueprint with restrictive filesystem and process defaults, credential filtering, routed inference, state integrity checks, and operator approval for blocked network endpoints.

Source Source

NVIDIA Corporation · project

NVIDIA OpenShell

1 claims+

Open-source runtime that isolates autonomous agents and enforces declarative filesystem, network, process, inference, and credential policies.

Official source
Agent Tool GovernanceAgent Execution Isolation
PREVENTAUDIT

OpenShell runs autonomous agents in isolated sandboxes and enforces out-of-process filesystem, network, process, inference, credential, and skill policies at binary, destination, method, and path level, with developer-approved policy updates and an audit trail of allow and deny decisions.

Source Source