Spec Steering Workflow
Use this skill when work is long, interruptible, multi-phase, or needs recovery in a new session.
Start or recommend a spec when any of these are true:
- - The task is likely to take more than 20 minutes.
- The task has more than one deliverable.
- The task needs staged execution.
- The task is likely to be interrupted.
- The task needs research plus execution.
- The task must survive across sessions.
- The user asks for plan first, then execution.
This skill manages execution state with workspace files:
- - INLINECODE0
- INLINECODE1
- INLINECODE2
Default workflow:
- 1. Read
steering/workflow.md and steering/preferences.md. - If resuming, read
handoff.md, then tasks.md, then meta.json. - Work only on the current batch.
- After each batch, or every 20-30 minutes, write a checkpoint.
- Before stopping, update
handoff.md and meta.json.
Valid checkpoint types are only:
- - INLINECODE10
- INLINECODE11
- INLINECODE12
Do not treat "working on it", "continue", or "in progress" as valid progress reports.
Use {baseDir}/scripts/specctl.py for these operations:
- - INLINECODE14
- INLINECODE15
- INLINECODE16
- INLINECODE17
- INLINECODE18
- INLINECODE19
- INLINECODE20
- INLINECODE21
Read references only as needed:
- -
{baseDir}/references/workflow-rules.md for trigger and lifecycle rules - INLINECODE23 for checkpoint requirements
- INLINECODE24 for resume order and stale-state rules
- INLINECODE25 for coexistence with other skills
- INLINECODE26 for file contracts and required fields
Use templates from {baseDir}/assets/templates/. Keep the skill lean: detailed rules belong in references/, file bodies belong in assets/templates/, and only execution state belongs in workspace specs/ and steering/.
规范引导工作流
当工作耗时较长、可中断、多阶段或需要在新会话中恢复时,使用此技能。
在以下任一情况成立时,启动或推荐规范:
- - 任务预计耗时超过20分钟。
- 任务包含多个可交付成果。
- 任务需要分阶段执行。
- 任务可能被中断。
- 任务需要研究加执行。
- 任务必须跨会话保持。
- 用户要求先规划后执行。
此技能通过工作区文件管理执行状态:
- - specs/active//
- specs/archive//
- steering/
默认工作流:
- 1. 读取steering/workflow.md和steering/preferences.md。
- 若为恢复会话,读取handoff.md,然后读取tasks.md,最后读取meta.json。
- 仅处理当前批次。
- 每批次完成后,或每20-30分钟,写入检查点。
- 停止前,更新handoff.md和meta.json。
有效的检查点类型仅限于:
不得将处理中、继续或进行中视为有效的进度报告。
使用{baseDir}/scripts/specctl.py执行以下操作:
- - init --title <标题> --kind <类型>
- checkpoint --type done|blocked|failed --batch <批次ID> --next <下一步操作>
- status
- resume
- validate
- archive
- set-status --status ready|review|completed [--phase <阶段>] [--next <下一步操作>]
- doctor
仅在需要时读取参考资料:
- - {baseDir}/references/workflow-rules.md:触发与生命周期规则
- {baseDir}/references/checkpoint-rules.md:检查点要求
- {baseDir}/references/recovery-rules.md:恢复顺序与过期状态规则
- {baseDir}/references/integration-rules.md:与其他技能的共存规则
- {baseDir}/references/template-contracts.md:文件契约与必填字段
使用{baseDir}/assets/templates/中的模板。保持技能精简:详细规则归入references/,文件内容归入assets/templates/,仅执行状态归入工作区specs/和steering/。