ralph-loop-yylo Skill
通过 Ralph 循环精确执行一个明确指派的 YYLO Ledger 任务,直至生成经过验证的排队提交。仅在用户明确要求 ralph-loop-yylo 时使用。
安装方式:把技能目录放入 ~/.claude/skills/(Claude Code)或在 claude.ai 设置中启用;也可复制右侧安装命令一键添加。
技能指令原文(SKILL.md)
Execute one assigned task in the Ralph loop
Read references/implement.md completely and follow it.
Stay within the assigned task. Do not select unrelated work, edit tasks.md, auto-tag releases, push, deploy, mutate production, or broaden scope because another issue is noticed. Record a bounded related Kanban follow-up when necessary.
Keep durable instructions concise and evidence-backed. Status belongs in the task response and runtime receipts, not AGENTS.md.
Controller checkpoints are best-effort local durability warnings after terminal metadata is durable. They never gate yy pi, yy task, yy merge, product commits, candidates, or releases.
User-facing Record results
After creation, update, discovery, or handoff, report the Record kind/profile,
actual immutable Record ID, and actual Ledger slug from the returned Record or a
native get readback. Never invent a slug from the title or confuse it with the ID.
If the selected API omits a field, say it is unavailable rather than fabricate it.
IDs remain authoritative for relations and lifecycle operations; slugs aid discovery.
Complete assigned request
Treat the following as the complete user-assigned request. Preserve task references and directives literally; resolve them only through the normal agent workflow.
$ARGUMENTS