The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run to determine the repository root using git rev-parse --show-toplevel. This is a standard and safe use of command execution for environment discovery.
[PRIVILEGE_ESCALATION]: The installation script sets executable permissions on the generated pre-commit hook (chmod +x). This is a necessary and standard step for git hooks to function and does not grant elevated system privileges.
[REMOTE_CODE_EXECUTION]: The skill contains no remote code execution patterns. It relies entirely on local scripts included in the skill package and standard system binaries (git, python).
[DATA_EXFILTRATION]: There are no network operations or external data transfer mechanisms. All scanning and reporting occur within the local environment.
[NO_CODE]: The skill uses pure Python with no external library dependencies, reducing the supply chain attack surface.

specstory-guard