prompt-library-for-startups
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches Model Context Protocol (MCP) servers and sample agents from trusted AWS organizations including
awslabsandaws-samples, as well as well-known developer tools likeAstral. - [COMMAND_EXECUTION]: Instructions direct the agent to perform project scaffolding and infrastructure deployment using CLI tools such as
cdk,terraform, andaws-cli. These procedures are governed by safety rules requiring user verification of change sets (e.g.,cdk diff) before deployment. - [SAFE]: The skill incorporates strong security guardrails, including explicit instructions to avoid wildcard IAM permissions, mandatory citation of source URLs, and a validation disclaimer requiring user review of all generated outputs. It also enforces the use of AWS Secrets Manager for managing sensitive configuration values.
Audit Metadata