Slice PRDs or plans into machine-checkable tracer-bullet issues via the loop workflow.
npx skills add https://github.com/kenlck/skills --skill loop-issues
Please help me install this Agent Skill. Skill: loop-issues Source: https://github.com/kenlck/skills/tree/main/ken-loop/skills/loop-issues Command: npx skills add https://github.com/kenlck/skills --skill loop-issues
Slice a PRD or plan into independently-grabbable tracer-bullet issues with a machine-checkability gate, enabling automated governance and progress tracking across teams.
Slice a PRD or plan into tracer-bullet issues ready for automated landing by an agent.
High-intent search queries and answers about installing and using this skill.
To turn a PRD into tracer-bullet issues, you slice the plan into independently-grabbable work items equipped with machine-checkable acceptance criteria, enabling automated governance and distributed progress tracking across teams.
Machine-checkable acceptance criteria are explicit gate conditions and runnable checks attached to sliced PRD issues, allowing unattended agents to automatically verify non-change guarantees and prevent regressions without human intervention.
Slicing a product plan requires breaking it into small, verifiable tracer-bullet slices and publishing them through the loop workflow, labeling each issue as AFK or HITL to direct autonomous or human-in-the-loop execution.
Yes, generating tracer issues requires the .loop/config.md framework to establish explicit gate criteria, runnable checks, and non-change guarantees to prevent regression and enable the loop workflow orchestration.
AFK labels direct work items to unattended agents for autonomous execution, while HITL labels flag issues requiring human-in-the-loop intervention, preserving traceability from PRD to issue during the loop workflow distribution.
nishantnaagnihotriCoordinate Gate 2 PRD handoffs and validation checks before design progression.
saitarrunConvert project plans and PRDs into structured issue tracker tickets.
oxieBreak down implementation plans into 20-30 testable tasks with acceptance criteria.
SatangThevalueRoute and decompose multi-profile work into Kanban tasks without executing them.
