The Agent Skills Directory

[SAFE]: The skill follows a strict conversational model, asking one question at a time and using predefined options to gather requirements. This structure minimizes the risk of unintended agent behavior.
[SAFE]: File system access is limited to reading local reference materials and templates, and writing the final specification to a designated '.kiro/specs/' directory. These actions are transparent and aligned with the skill's discovery purpose.
[SAFE]: The use of the internal command '/kiro:spec-init' is a standard workflow integration and does not involve arbitrary shell execution or external script invocation.
[SAFE]: No obfuscation, hardcoded credentials, or external network requests are present in the skill code or metadata.

intent-explorer