>
    Strata Copilot
    Prisma AIRS Integration with Cortex AES for AI Coding Agent Security
    Focus
    Download PDF
    Focus
    1. Home
    2. Prisma AIRS
    3. AI Runtime Security
    4. AIRS API
    5. Prisma AIRS Integration with Cortex AES for AI Coding Agent Security
    Download PDF
    Prisma AIRS

    Prisma AIRS Integration with Cortex AES for AI Coding Agent Security

    Table of Contents

    Prisma AIRS Integration with Cortex AES for AI Coding Agent Security

    Set up Prisma® AIRS™ in Strata Cloud Manager so that Cortex Agentic Endpoint Security (AES) can enforce AI security policies on AI coding agents running on developer endpoints.
    Where Can I Use This?What Do I Need?
    • Prisma AIRS — AI Runtime Security (API Intercept)
    • Prisma AIRS AI Runtime API application
    • Cortex AES subscription
    Prisma AIRS integrates with Cortex Agentic Endpoint Security (AES), an endpoint governance platform, to secure AI coding tools such as Claude Code, Cursor, Codex, GitHub Copilot, and Antigravity on developer endpoints. Prisma AIRS inspects the content flowing through those agents — prompts, MCP tool calls, and network calls — and returns an enforcement verdict. AES owns everything around that verdict on the endpoint: it decides which agents, endpoint groups, and events are in scope, deploys and manages the AIRS hooks on every device with API keys handled centrally, enforces the verdict and informs the developer inside the agent's chat, and records every verdict in the AES portal attributed to the device, agent, session, and policy. Detection rules are defined once in the Prisma AIRS console in Strata Cloud Manager and pushed to thousands of endpoints as managed policy, with no per-device hook setup.
    Prisma AIRS inspects user prompts before they reach the LLM and after the response returns, external MCP tool calls before execution, and network calls (WebFetch and WebSearch) before and after execution. It detects prompt injection, sensitive data leakage, insecure output such as malware and malicious URLs, toxic content, and credential leakage. When sensitive data is detected, the response returns masked text in place of the original content.
    You define and manage security profiles entirely in the Prisma AIRS console in Strata Cloud Manager. A profile specifies which detectors run and their configured actions — for example, prompt injection set to block. Once connected, AES fetches your security profiles so a runtime policy can reference them by name: a default profile applies to the whole policy, and any individual rule for prompts, MCP tools, or network calls can override it with a different profile. To enable this integration, set up Prisma AIRS in Strata Cloud Manager and then complete the remaining configuration in the AES portal.

    Set Up Prisma AIRS for the AES Integration

    Complete the following steps in Strata Cloud Manager before you configure the integration in the AES console. All configuration specific to AES — connecting the integration, creating a runtime policy, and deploying hooks to endpoints — is covered in the AES documentation. The only value you copy between the two consoles is the service account credential; AES uses it to fetch your API keys and security profiles.
    1. Create a service account in Strata Cloud Manager to generate the credentials AES uses to connect to Prisma AIRS.
      1. Add a Service Account.
      2. Enter a name for the service account.
      3. Assign the RBAC role required for Prisma AIRS API access. Minimum permissions required are Read Only roles for API Keys and Profiles roles.
      4. Click Save and copy the Client ID and Client Secret.
      These credentials are displayed only once. Store them securely before continuing — you will enter them in the AES console in a later step.
    2. Create a runtime API application and generate an API key to use across scans for a given policy.
      For more information, see Manage Applications, API Keys, Security Profiles, and Custom Topics for steps to create the application and generate the key.
      Create one API key per policy. If you plan to apply different security profiles across policies, create a separate API key for each.
    3. Create a security profile to define which detectors run on events that AES forwards to Prisma AIRS.
      See Create and Configure API Security Profile for steps to configure detectors — for example, set prompt injection to block, or add data loss prevention and malicious URL detectors.
      If an AES policy targets more agents than the selected security profile covers, those agents will not be fully protected. Scope the security profile and the AES policy to the same set of agents to ensure consistent coverage.
    4. Complete the integration setup in the AES console using the Client ID and Client Secret from Step 1.
      AES uses the service account credentials to fetch your API keys and security profiles automatically. For all remaining configuration steps, see the AES documentation.

    © 2026 Palo Alto Networks, Inc. All rights reserved.