aws-sst-development

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidance for using the SST v4 framework, incorporating security-conscious practices such as verifying AWS identity with aws sts get-caller-identity before deployment and cautioning the user to clean up state files that may contain sensitive account details.
  • [COMMAND_EXECUTION]: The skill documents the use of industry-standard CLI tools including npx sst, npx vitest, and the aws CLI. These tools are appropriate and necessary for the skill's stated purpose of managing infrastructure as code.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze project-specific files such as sst.config.ts and resources in the infra/ directory. While this represents a surface for indirect prompt injection if those files were to contain malicious instructions, this behavior is fundamental to the skill's function as a code expert and contains no specific exploitable patterns.
  • [DYNAMIC_EXECUTION]: The skill operates within the SST/Pulumi ecosystem, which inherently uses dynamic execution of TypeScript to generate and apply infrastructure plans. This is the intended behavior of the framework being documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 10:09 AM
Security Audit — agent-trust-hub — aws-sst-development