The Agent Skills Directory

[Prompt Injection] (SAFE): The instructions are focused on cloud architecture tasks and do not contain any commands to bypass safety filters, reveal system prompts, or ignore previous instructions.- [Data Exposure] (SAFE): There are no hardcoded credentials, API keys, or attempts to access sensitive local system files.- [Remote Code Execution] (SAFE): No external scripts are downloaded or executed. The skill does not contain any code-based components.- [Indirect Prompt Injection] (LOW): The skill is designed to process user-provided architecture requirements. While it generates Terraform code based on these inputs, it does not autonomously execute the code or perform network requests, limiting the risk to the generation of potentially insecure configurations which require user review.- [Persistence] (SAFE): No mechanisms for maintaining access or modifying system startup configurations were detected.- [No Code Presence] (INFO): The skill consists entirely of markdown instructions without any associated scripts or executables, which minimizes the technical threat profile.

cloud-architect