The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a Python script (scripts/docs_inventory.py) that invokes git via subprocess.run to gather file history and repository metadata. These calls use the list-based argument format and do not invoke a shell, following secure coding practices.
[SAFE]: The skill provides a proactive security checklist that instructs the agent to use ripgrep to search for accidentally committed secrets, passwords, and API keys within the documentation tree.
[SAFE]: Structural modifications in 'restructure' mode are governed by a migration ledger and require explicit user approval before execution, mitigating the risk of accidental data loss or unauthorized file changes.
[EXTERNAL_DOWNLOADS]: The documentation references well-known and trusted third-party tools for documentation quality assurance, such as lychee for link checking and Vale for prose linting. These are standard development tools and do not involve untrusted remote code execution.

docs-architect