ingesting-into-data-lake
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill instructions are focused on technical workflows for AWS data ingestion. No patterns suggesting prompt injection, safety filter bypasses, or instruction overrides were detected.
- [DATA_EXFILTRATION]: No hardcoded credentials (such as API keys or secrets) were found. The skill utilizes AWS Glue connections and Secrets Manager placeholders for managing database credentials, which is a secure practice. There is no evidence of unauthorized data transfer to non-whitelisted domains.
- [REMOTE_CODE_EXECUTION]: The skill generates and executes PySpark scripts for AWS Glue jobs. These scripts are built from predefined, safe templates included in the skill's reference documentation and are used for legitimate ETL tasks within the user's controlled AWS environment.
- [COMMAND_EXECUTION]: The skill uses AWS CLI commands for administrative operations like listing S3 objects or starting Glue crawlers. These commands are standard for the described use case and do not involve executing arbitrary or unsanitized user input.
- [SAFE]: The skill adheres to established security best practices for data engineering within AWS. The source repository is a well-known organization, and the workflows include mandatory validation steps for row counts and data quality.
Audit Metadata