The Agent Skills Directory

[SAFE]: The skill is restricted to read-only operations using the Read, Glob, and Grep tools. It does not have access to tools that could modify files or execute arbitrary system commands.
[SAFE]: No network tools are enabled, preventing any potential for data exfiltration. The skill analyzes local project documentation such as sprint plans, ADRs, and GDD requirements.
[SAFE]: The skill processes content from local story and ADR files. While these files are external inputs to the skill's logic, the execution environment is constrained to validation and a controlled task hand-off to a secondary agent (qa-lead) via a predefined gate mechanism, which is a safe operational pattern.

story-readiness