migrating-to-amazon-redshift

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Dynamic Script Generation: The skill relies on the AI generating execution scripts (Bash and Python) at runtime based on provided knowledge templates. To manage the potential risks of dynamic code, the skill includes explicit instructions for the AI to validate and sanitize inputs—specifically database identifiers—using regular expressions to prevent injection attacks.
  • External Package Management: Generated scripts are instructed to install necessary Python libraries such as boto3 and teradatasql. These are standard, well-known packages used for cloud and database connectivity, and their use is consistent with the skill's intended administrative purpose.
  • Input Sanitization and Indirect Injection Surface: The skill ingests metadata from source databases which could potentially be used as a vector for indirect prompt injection. The provided instructions mitigate this by requiring the AI to validate every identifier against a strict allowlist pattern (^[A-Za-z_][A-Za-z0-9_]*$) before use in generated SQL or shell commands.
  • Adherence to Security Best Practices: The skill proactively guides the user toward secure configurations, including the use of secret management services for credential storage, enforcement of TLS for data in transit, and the use of IAM roles for least-privilege authorization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 07:58 AM
Security Audit — agent-trust-hub — migrating-to-amazon-redshift