Panshi
EN/

Home / Guardrails & Safety

LLM Guard (Protect AI)

Open-source input/output scanner toolkit (35+ scanners) for PII, injection and toxicity checks on LLM traffic.

CategoryGuardrails & Safety
Open sourceYes
Self-hostableYes
Pricing modelfree
Pricing notesMIT OSS scanners free; enterprise via Palo Alto Networks (Protect AI acquisition)
Framework integrationsopenai-sdk, langchain
Funding / ownershipProtect AI acquired by Palo Alto Networks (2025)
GitHub stars3,068
Maintenanceslow (2025-12-15)
License (GitHub)MIT
Open issues35
Maturity signal64/100 (Established) — computed from public GitHub signals, see methodology

Pricing/feature source: https://github.com/protectai/llm-guard

Maturity signal is computed from public GitHub data only. How it is calculated.

Frequently asked questions

Is LLM Guard (Protect AI) open source?

Yes. Licensed under MIT.

Can I self-host LLM Guard (Protect AI)?

Yes — LLM Guard (Protect AI) can run in your own infrastructure.

How much does LLM Guard (Protect AI) cost?

Free. MIT OSS scanners free; enterprise via Palo Alto Networks (Protect AI acquisition)