aws-helper
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill suggests that users manually install an external extension from an unverified GitHub repository (
github:mindverse/skillhub) when the user requests support for other cloud platforms such as Azure or GCP. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to generate complex AWS infrastructure configurations and IAM policies, representing a potential surface for injection attacks.
- Ingestion points: User-provided business and technical requirements for cloud architecture (SKILL.md).
- Boundary markers: The instructions do not define clear delimiters or use specific markers to prevent the agent from obeying instructions embedded within the user's requirements.
- Capability inventory: The skill is restricted to generating text-based templates (JSON, YAML, TypeScript, Python) and explicitly states it cannot execute commands or access AWS accounts, which limits the impact of potential injections.
- Sanitization: There is no evidence of sanitization or validation performed on the user requirements before they are used to generate infrastructure-as-code templates.
Audit Metadata