The Agent Skills Directory

[SAFE]: The skill implements a logical interaction flow for requirements gathering and pressure-testing project designs without any malicious instructions.
[DATA_EXPOSURE]: The skill performs local codebase exploration using search tools (Grep, Glob) to gather context for its questions. This data access is confined to the local project environment and is necessary for the skill's stated purpose of technical clarification.
[COMMAND_EXECUTION]: The skill interacts with the local file system to write interview summaries into the docs/idea/ directory. These operations use restricted paths and are part of the core functionality requested by the user.

ito-grill