The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a test script (scripts/test_skill.py) that uses subprocess.run to execute other internal scripts (validate.py, analyze_app_resilience.py) for validation purposes. These calls use sys.executable and target files within the skill's own directory structure, which is a standard practice for development tools.
[DATA_EXPOSURE]: The instructions in references/observability.md and SKILL.md explicitly emphasize security and privacy, instructing the agent to never log secrets, tokens, PII, or credentials. This aligns with security best practices.
[REMOTE_CODE_EXECUTION]: The scripts/analyze_app_resilience.py tool performs static analysis on local application code using regular expressions to detect common resilience issues. It does not download or execute remote code, and its operations are restricted to reading files from a path provided by the user.

maintainable-app