aws-cost-operations
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from external AWS sources, creating a surface for indirect prompt injection.
- Ingestion points: Data is retrieved via tools like mcp__cw__*, Bash(aws logs *) for CloudWatch logs, and Bash(aws cloudtrail *) for audit activity.
- Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore potentially malicious content embedded within logs or billing reports.
- Capability inventory: The skill possesses extensive capabilities via Bash(aws *) tools and multiple MCP servers to query, monitor, and manage AWS infrastructure.
- Sanitization: No evidence of sanitization or filtering logic is present to handle retrieved external data before it is processed by the agent.
Audit Metadata