The Agent Skills Directory

[SAFE]: The skill implements a task-breakdown workflow that converts project specs into vertical-slice issues. It includes a mandatory human-in-the-loop validation step where the agent must present the proposed breakdown and dependency graph for user approval before publishing to an issue tracker. While the skill reads external data (issue bodies and comments), the risk of indirect prompt injection is mitigated by this required user confirmation. The functionality is limited to reading and creating issues within a project's existing infrastructure, which is consistent with the stated purpose.

to-issues