amazon-nova-reel

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides detailed instructions and code for using Amazon Bedrock Nova Reel, focusing on official AWS features and documentation.\n- [COMMAND_EXECUTION]: The included Python script performs legitimate operations such as AWS API calls via the boto3 library and local file I/O for state tracking using secure file permissions (0o600).\n- [EXTERNAL_DOWNLOADS]: The skill references multiple official AWS documentation URLs for model lifecycle, pricing, and API usage. These are verified as trusted sources and do not involve executable code downloads.\n- [DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration. The script interacts exclusively with user-configured AWS S3 buckets and Bedrock endpoints within the user's AWS account.\n- [CREDENTIALS_UNSAFE]: The skill does not contain hardcoded credentials and correctly instructs the operator to use environment variables for sensitive configuration.\n- [PROMPT_INJECTION]: No prompt injection or override patterns were identified in the instructions or the code.\n- [SAFE]: The provided automation script implements advanced safety features, such as SHA-256 request integrity verification, explicit dry-run gating, and S3 URI prefix validation to prevent accidental or malicious usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 01:52 PM
Security Audit — agent-trust-hub — amazon-nova-reel