hf-cloud-sagemaker-production-defaults

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The deployment and teardown scripts use subprocess.run to execute AWS CLI commands for infrastructure management. The implementation uses list-based arguments rather than shell strings, which is a security best practice that prevents shell injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: The skill instructions describe downloading model artifacts from trusted sources, including user-controlled Amazon S3 buckets and the Hugging Face Hub. These operations are standard for SageMaker deployments and do not involve executing untrusted remote code.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by processing user-supplied inputs such as model names and environment variables. However, these inputs are passed to parameterized AWS API calls via boto3 or safe CLI calls, which effectively mitigates the risk of prompt or command injection.
  • [SAFE]: No malicious patterns, such as hardcoded credentials, data exfiltration, or code obfuscation, were detected in the skill scripts or instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 06:18 AM
Security Audit — agent-trust-hub — hf-cloud-sagemaker-production-defaults