The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the trailmark and find utilities via the shell to perform structural analysis on a target directory.
Evidence: find {args} -type f ... and trailmark analyze --summary {language_flag} {args} in the Execution section.
Note: This is standard functionality for a command-line wrapper skill. While {args} is interpolated into the shell command, the skill's purpose is to analyze user-specified local directories.
[SAFE]: The skill implements strict environment constraints by explicitly forbidding the agent from performing automated installations.
Evidence: "Do NOT run pip install, uv pip install, git clone, or any install command. The user must install trailmark themselves."
This prevents the agent from introducing unauthorized or unverified external software into the host environment.

trailmark-summary