aid: protect-ai
name: Protect AI
description: >-
  Protect AI is an AI/ML security company building defensive tooling and
  open-source security research for the model supply chain, large language
  model (LLM) applications, and runtime AI workloads. The commercial platform
  is anchored by three products — Guardian (AI model security and
  third-party model scanning across PyTorch, TensorFlow, ONNX, Keras, Pickle,
  GGUF, Safetensors, and 35+ formats with Hugging Face, SageMaker, MLflow,
  S3, Git, and Artifactory integrations), Recon (scalable AI red teaming with
  an attack library of 450+ known attacks against LLM applications mapped to
  OWASP LLM Top 10), and Layer (runtime AI security with 27 turnkey policies,
  eBPF and SDK collection, and DataDog/Splunk/Elastic/PagerDuty integration).
  Protect AI also stewards a portfolio of widely used open-source AI security
  projects on GitHub including LLM Guard (input/output scanner toolkit for
  prompt injection, PII, toxicity, secrets, and data leakage detection),
  ModelScan (model serialization attack scanner), Rebuff (prompt injection
  detector, archived May 2025), NB Defense (Jupyter notebook security), AI
  Exploits (real-world AI/ML vulnerability demonstrations), and Vulnhuntr
  (LLM-driven autonomous vulnerability discovery), plus the huntr bug bounty
  platform for AI/ML with over 17,000 security researchers. Protect AI was
  acquired by Palo Alto Networks in 2025 and is being integrated into the
  Prisma AIRS AI security platform. No public, programmatic developer API or
  OpenAPI specification is published for Guardian, Recon, or Layer at the
  time of this profile — integration is performed via product CLIs, SDKs,
  eBPF agents, and platform connectors documented behind Palo Alto Networks
  customer accounts.
type: Index
position: Provider
access: 3rd-Party
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
tags:
  - AI Security
  - ML Security
  - MLSecOps
  - AI Model Security
  - AI Red Teaming
  - LLM Security
  - Prompt Injection
  - Model Scanning
  - Runtime Security
  - Supply Chain Security
  - AI Bug Bounty
  - Open Source
  - Acquired
  - Palo Alto Networks
url: https://raw.githubusercontent.com/api-evangelist/protect-ai/refs/heads/main/apis.yml
created: '2026-05-25'
modified: '2026-05-25'
specificationVersion: '0.20'
apis: []
common:
  - type: DomainSecurity
    url: security/protect-ai-domain-security.yml
  - type: Website
    url: https://protectai.com
  - type: Product
    url: https://protectai.com/guardian
  - type: Product
    url: https://protectai.com/recon
  - type: Product
    url: https://protectai.com/layer
  - type: Blog
    url: https://protectai.com/blog
  - type: Newsroom
    url: https://protectai.com/newsroom
  - type: Training
    url: https://protectai.com/course-mlsecops-foundations
  - type: Documentation
    url: https://protectai.com/secure-by-design
  - type: BugBounty
    url: https://huntr.com
  - type: ParentCompany
    url: https://www.paloaltonetworks.com
  - type: ParentProduct
    url: https://www.paloaltonetworks.com/prisma/airs
  - type: GitHubOrganization
    url: https://github.com/protectai
  - type: OpenSource
    name: LLM Guard
    url: https://github.com/protectai/llm-guard
  - type: OpenSource
    name: ModelScan
    url: https://github.com/protectai/modelscan
  - type: OpenSource
    name: Rebuff (archived)
    url: https://github.com/protectai/rebuff
  - type: OpenSource
    name: NB Defense
    url: https://github.com/protectai/nbdefense
  - type: OpenSource
    name: AI Exploits
    url: https://github.com/protectai/ai-exploits
  - type: OpenSource
    name: Vulnhuntr
    url: https://github.com/protectai/vulnhuntr
  - type: Documentation
    name: LLM Guard Documentation
    url: https://llm-guard.com
  - type: LinkedIn
    url: https://www.linkedin.com/company/protect-ai
  - type: Twitter
    url: https://twitter.com/ProtectAICorp
  - type: YouTube
    url: https://www.youtube.com/@protectai
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com