The Agent Skills Directory

[SAFE]: The skill implements a product management workflow based entirely on local markdown templates. It does not perform any network operations or access external resources.
[SAFE]: User interaction is limited to predefined subcommands (problem, competitive, feasibility, checkpoint). The instructions include clear logic for parsing these arguments and showing a usage error for unrecognized inputs, which prevents directory traversal or unauthorized file access.
[SAFE]: No sensitive data, such as API keys or environment variables, is accessed, required, or stored by the skill.
[SAFE]: All tool usage (Read, Write, Grep, Glob) is consistent with the skill's purpose of documentation management, and no dynamic code execution or privilege escalation patterns were detected.

discovery