Prompt Security

What is Prompt Security?

Prompt Security is a generative AI security product, now branded as from SentinelOne, that protects organizations using LLM applications and AI agents. It applies guardrails for policy, topic detection, sensitive data, secrets and prompt injection.

Key features of Prompt Security

• Guardrails: policy, topic, sensitive data, secrets and prompt injection controls with a usage dashboard.
• OneClaw: tracks and analyzes OpenClaw agent deployments across an organization, with visibility into activity and usage.
• ClawSec: a security skill suite for OpenClaw, NanoClaw and Hermes agents, with drift detection, automated audits and skill integrity verification.
• Prompt Fuzzer (ps-fuzz): an open source tool that tests system prompt resilience against LLM attacks, with 16 providers and 16 attack types.

How Prompt Security works

Prompt Security sits between users or applications and AI models, inspecting prompts and responses against policy. Security teams view usage and violations in a dashboard, and agent-specific tools monitor autonomous agent deployments.

Who uses Prompt Security?

It is used by security teams at organizations rolling out copilots, internal LLM apps and AI agents that need visibility and control over AI usage.

Prompt Security pricing

Prompt Security does not publish pricing; visitors are directed to request a demo.

Prompt Security integrations

The vendor site does not describe product integrations; the open source Prompt Fuzzer supports 16 LLM providers.

Prompt Security alternatives

Protect AI focuses on ML model and supply chain security. Lakera offers real-time guardrails against prompt injection and data leakage for LLM apps. HiddenLayer provides model scanning and runtime defense for machine learning assets.