aws-well-architected
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
membraneCLI to manage connections, discover actions, and perform API requests. This is the intended interface for interacting with the Membrane platform. - [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@membranehq/clipackage from the official NPM registry, which is a standard procedure for this vendor's tooling. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external sources, specifically the AWS Well-Architected API and Membrane's action library. This creates an attack surface where instructions embedded in external metadata or API responses could influence agent behavior. However, the skill uses structured JSON output and specific tool parameters which helps reduce the risk of accidental command obedience.
Audit Metadata