The Agent Skills Directory

[SAFE]: The skill employs standard, read-only commands using git and the GitHub CLI (gh) to gather metadata about the project's current status.
[SAFE]: It retrieves workflow-related state from a local directory (~/.claude/research/) used to store markdown research artifacts generated by the skill suite.
[SAFE]: The process instructions include strong safety constraints, explicitly forbidding the AI from automatically executing any suggested actions and requiring user intervention for the next step.
[SAFE]: All external documentation links and repository references point to the author's own infrastructure, and no evidence of malicious behavior or data exfiltration was detected.

help