Focus

Prisma AIRS

Scanning Models

Table of Contents

Organize Security Scans with Custom Labels

Scanning Models

Scan a Hugging Face model, local model, or object storage model using CLI/SDK.

Where Can I Use This?	What Do I Need?
Prisma AIRS (AI Model Security)	Prisma AIRS AI Model Security License

Once your Security Group is configured, you can scan models through either the CLI or Python SDK. The process varies slightly depending on whether you're scanning Hugging Face AI models or local models.

While scanning a model using Python SDK:

you will need to use ModelSecurityAPIClient which is the base object to perform API calls.
you can configure the base_url using environment variables or in your code.

When you scan using SDK, it's your responsibility to enforce allow or block decisions according to the scan evaluation outcomes.

When you scan using CLI, the CLI will exit with a non-zero exit code if the model is unsafe.

AI Model Security can handle up to 1,000 files per scan.
You cannot delete a scan.

Scan a Hugging Face Model

Scan a model hosted on Hugging Face.

Scan a Hugging Face Model

To scan a model hosted on Hugging Face, provide the model URI and your security group UUID. For Hugging Face AI models, only model_uri is required.

Before you start the model scanning process:

Ensure that the security group source type must match the source of the model that you are scanning. For example, you cannot use a S3 security group on a Hugging Face.
Verify and ensure that HuggingFace.co domain (https://huggingface.co/) is allowed.
Ensure that the ignore_patterns and allow_patters do not overlap with each other.

We don’t support private Hugging Face repositories. You can only scan public Hugging Face repositories. If you want to scan private Hugging Face repository, then you can download the model and scan it using local model scan.

When creating a scan, you can attach up to 50 custom labels to help organize your scans.

Scan using CLI

model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \
  --model-uri "https://huggingface.co/microsoft/DialoGPT-medium" \
  -l env=production

Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims"
)

result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    model_uri="https://huggingface.co/microsoft/DialoGPT-medium",
    labels={ "env": "production" }
)

print(f"Scan completed: {result.eval_outcome}")

The AI Model Security automatically fetches the latest version from Hugging Face. To scan a specific version, include the version parameter.

Scan using CLI


model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \
  --model-uri "https://huggingface.co/microsoft/DialoGPT-medium" \
  --model-version "7b40bb0f92c45fefa957d088000d8648e5c7fa33"

Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims"
)
result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    model_uri="https://huggingface.co/microsoft/DialoGPT-medium",
    model_version="7b40bb0f92c45fefa957d088000d8648e5c7fa33"
)

Filter Files in Hugging Face Scans

Large Hugging Face repositories may contain files you don't need to scan. Use global patterns to include or exclude specific files.

Scan using CLI


model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \
  --model-uri "https://huggingface.co/microsoft/DialoGPT-medium" \
  --allow-patterns "*.bin" "*.json" \
  --ignore-patterns "*.md" "*.txt"

Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims"
)
result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    model_uri="https://huggingface.co/microsoft/DialoGPT-medium",
    allow_patterns=["*.bin", "*.json"],
    ignore_patterns=["*.md", "*.txt"]
)

Scan a Local Model

Scan a model that is stored locally.

Scan a Local Model

For models stored locally, specify the path to the model directory. To scan a model from the local disk, only model_path is required.

Before you start the model scanning process:

Ensure that the security group source type must match the source of the model that you are scanning. For example, you cannot use a Hugging Face security group on a local model. If you don’t provide any model URI, then by default local disk source type is used.
Validate that the model path points to the correct storage location.
The ignore_patterns and allow_patters is not applicable for local model scans.
Running a model scan can consume up to 4GB memory depending on the size and type of the model. Therefore, ensure that the environment used for the scanning has sufficient resources. Verify if you've enough space to download and sàve the model being scanned.
Use model_path to specify the local disk location for models.

When creating a scan, you can attach up to 50 custom labels to help organize your scans.

Scan using CLI


model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \
  --model-path "path/to/local/model" \
  -l env=production

Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims"
)
result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    model_path="path/to/local/model",
    labels={ "env": "production" }
)

Scan a Model from Object Storage

Scan a model from object storages like Amazon S3, Google Cloud Storage, Azure Blob Storage, JFrog Artifactory, and Gitlab Model Registry.

Scan a Model from Object Storage

We support object storages Amazon S3, Google Cloud Storage, Azure Blob Storage, JFrog Artifactory, and Gitlab Model Registry. To scan an AI model from these cloud storage models, provide the URL of these models as model_uri parameter while calling the scan on the SDK.

The model security SDK will perform the download for you and queue the model for scan.

When creating a scan, you can attach up to 50 custom labels to help organize your scans.

Scan using CLI


model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \

  --model-uri "<model_uri>" \
  --model-name "production-classifier" \
  --model-author "ml-team" \
  --model-version "v2.1" \
  -l env=production

Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims"
)
result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    
    model_uri="<model_uri>",
    model_name="production-classifier",
    model_author="ml-team",
    model_version="v2.1",
    labels={ "env": "production" }
)

The model_uri parameter must use the format of supported cloud storage platforms:

Amazon S3 (s3://)
Google Cloud Storage (gs://)
Azure Blob Storage (https://[account].blob.core.windows.net/)
JFrog Artifactory (https://[instance].jfrog.io/)
GitLab Model Registry (https://[gitlab-instance]/-/ml/models/)

The CLI shows scan results in real-time as they finish. Each scan tests the model against all active rules in your Security Group. The output shows whether the model passes or fails based on your rule configuration.

A model fails if any blocking rule detects a violation. Non-blocking rules record findings without preventing the model from being approved.

Customize Model Scans

Customize your AI model scans.

Customize Model Scans

You can configure scan execution and adjust result timeout settings.

Customize Scan using CLI


model-security scan \
  --security-group-uuid "12345678-1234-1234-1234-123456789012" \
  --model-uri "<model_uri>" \
  --poll-interval-secs 10 \
  --poll-timeout-secs 900 \
  --download-timeout-secs 1800 \             # Object storage download timeout
  --download-dir "/custom/download/path"\    # Object storage download location
  --cleanup-download-dir \			# Cleanup downloads after scan
  --block-on-errors

Customize Scan using Python SDK


from model_security_client.api import ModelSecurityAPIClient

# Initialize the client with download configuration
client = ModelSecurityAPIClient(
    base_url="https://api.sase.paloaltonetworks.com/aims",
    download_timeout_secs=1800,            # Object storage download timeout
    download_dir="/custom/download/path",  # Object storage download location
    cleanup_download_dir=True              # Cleanup downloads after scan
)

# Perform scan with polling configuration
result = client.scan(
    security_group_uuid="12345678-1234-1234-1234-123456789012",
    model_uri="<model_uri>",
    poll_interval_secs=10,
    poll_timeout_secs=900,
)

Following are the configuration options to customize the scan for AI models.

Configuration Option	Description	Default Value
download_timeout_secs	(`Object storage scans only`) Specify the timeout duration for model downloads from cloud storage.	600 seconds
download_dir	(`Object storage scans only`) Specify the destination directory for downloading models from object storage.	~/.cache/airsms/
cleanup_download_dir	(`Object storage scans only`) Remove downloaded models after scanning to conserve disk space.	False
poll_interval_secs	Specify the frequency of scan status checks.	5 seconds
poll_timeout_secs	Specify the maximum wait time for scan completion.	600 seconds

block_on_errors	(`CLI only`) CLI exits with an error code when scan errors occurs.	NA

Customizing Security Groups

Organize Security Scans with Custom Labels

Prisma AIRS Docs

Scanning Models

Prisma AIRS Docs

Scanning Models

Scan a Hugging Face Model

Scan a Hugging Face Model

Filter Files in Hugging Face Scans

Scan a Local Model

Scan a Local Model

Scan a Model from Object Storage

Scan a Model from Object Storage

Customize Model Scans

Customize Model Scans

Activation & Onboarding

Next-Generation Firewalls

Cloud-Delivered Security Services

Network Security

Visibility & Monitoring