The Agent Skills Directory

[SAFE]: The skill provides a suite of static analysis scripts for architectural and security auditing. These scripts perform read-only operations on local project files to generate reports on system design and security posture.
[SAFE]: File processing is handled securely using the pathlib module for file system traversal and yaml.safe_load for configuration files. Static analysis of Python source code is performed via the ast (Abstract Syntax Trees) module, which inspects code structure without executing it.
[INDIRECT_PROMPT_INJECTION]: The skill processes external files (code, configurations, and dependencies) provided by the user, creating an indirect prompt injection surface. However, the analysis is limited to keyword detection and structural parsing, and the skill lacks network or shell execution capabilities, which mitigates the risk of malicious instructions being acted upon.
[SAFE]: No hardcoded credentials, sensitive file access (such as SSH keys or AWS credentials), or network exfiltration patterns were detected in the scripts. The subprocess module is imported in several scripts but is not utilized to execute external commands.

architect-reviewer