The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes subprocess.run and subprocess.Popen to execute local commands for Git repository inventory and external analysis tools (e.g., ast-grep, semgrep, cargo). These executions are performed with shell=False, ensuring that shell interpolation risks are avoided. The tools are invoked for the legitimate purpose of architectural auditing.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code from the repository being audited. While this presents a surface for indirect prompt injection, the skill acts as a reporting and enforcement tool rather than a generic instruction follower for the data it processes. No evidence of unsafe interpolation was found.

architecture-enforce