aws-bedrock-evals

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • NO_CODE (SAFE): The analyzed files (README.md, metadata.json) do not contain a SKILL.md or any executable logic. As a result, there is no code behavior to evaluate for malicious patterns.
  • EXTERNAL_DOWNLOADS (LOW): The documentation mentions an installation source (antstackio/skills) that is not on the pre-approved trusted list. Users are advised to verify the integrity of the repository before running the installation command.
  • INDIRECT_PROMPT_INJECTION (LOW): The skill's documentation identifies an attack surface where it processes untrusted user data (system prompts and test scenarios) and possesses high-privilege capabilities (IAM role creation and AWS CLI execution). However, no implementation code exists to assess specific vulnerabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 07:28 AM