rds-operation-review
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes data from external sources, specifically RDS engine logs and event streams, to identify operational issues.
- Ingestion points: Data is retrieved via
logs.FilterLogEvents(SKILL.md Step 6) andrds.DescribeEvents(SKILL.md Step 7). - Boundary markers: The instructions do not currently include specific delimiters or directives to the agent to disregard potential instructions that might be embedded within retrieved log entries.
- Capability inventory: The agent has access to various AWS read-only APIs and the ability to generate Markdown artifacts as part of its reporting function.
- Sanitization: No explicit sanitization of log data is mentioned before the data is evaluated or included in the final report.
- Extensive AWS API Usage: The skill instructions involve a significant number of AWS API calls to gather configuration, performance, and cost data. These operations are aligned with the skill's stated purpose of conducting a thorough database review.
- Read-Only Posture: The skill is structured to perform data collection and analysis only. It does not include commands for resource modification or deletion, supporting a secure-by-design auditing approach.
Audit Metadata