aws-security-audit
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads and executes MCP servers from the AWS Labs repository using the
uvxtool. This is a well-known source for official AWS experimental and open-source tooling. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from AWS CloudTrail API events. While this data is externally generated, the skill's capabilities are scoped to read-only operations, and it lacks the ability to write files or execute arbitrary code based on the log content.
- [CREDENTIALS_UNSAFE]: The skill instructions correctly identify the need for AWS credentials (
AWS_ACCESS_KEY_ID,AWS_SECRET_ACCESS_KEY) and appropriately suggests providing them via environment variables, which is standard safe practice for secret management.
Audit Metadata