The Agent Skills Directory

[SAFE]: The skill operates within a well-defined scope, interacting primarily with project documentation in the docs/ and design/ directories. It utilizes the Read, Glob, and Grep tools to gather context from existing ADRs and engine references, which is essential for its primary function.
[SAFE]: Modification of files (using Write and Edit) and updates to the architectural registry are strictly gated by user confirmation via the AskUserQuestion tool, ensuring human oversight for all repository changes.
[SAFE]: The use of sub-agents for technical reviews (Engine Specialist and Technical Director) follows a structured internal protocol and does not involve untrusted external execution or network activity.

architecture-decision