/pace-trace
Use when user asks "why did devpace decide X", "追溯", "为什么这样做", "决策记录", "决策原因", "架构决策", "ADR", "技术选型", wants to see AI decision trail or manage architecture decisions, or says /pace-trace [CR] [gate/decision/arch]. NOT for concept explanations or methodology questions (use
$ npx -y skills add arch-team/devpace --skill pace-trace --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/pace-trace
Context preview
The summary Claude sees to decide when to auto-load this skill.
Use when user asks "why did devpace decide X", "追溯", "为什么这样做", "决策记录", "决策原因", "架构决策", "ADR", "技术选型", wants to see AI decision trail or manage architecture decisions, or says /pace-trace [CR] [gate/decision/arch]. NOT for concept explanations or methodology questions (use
SKILL.md
pace-trace.SKILL.mddescription: Use when user asks "why did devpace decide X", "追溯", "为什么这样做", "决策记录", "决策原因", "架构决策", "ADR", "技术选型", wants to see AI decision trail or manage architecture decisions, or says /pace-trace [CR] [gate/decision/arch]. NOT for concept explanations or methodology questions (use /pace-theory). NOT for metrics or retrospective analysis (use /pace-retro).
allowed-tools: AskUserQuestion, Read, Write, Edit, Glob, Grep
argument-hint: "[CR 名称或编号] [gate1|gate2|gate3|intent|change|risk|autonomy|timeline|arch]"
model: haiku
/pace-trace — 决策轨迹查询
> 三层渐进透明的深入层。提供完整的 AI 决策轨迹,让用户可审计 Claude 的每个系统判断。
与相关功能的区别
| 功能 | 回答的问题 | 数据维度 | |------|-----------|---------| | `/pace-trace [CR] [type]` | "Claude 为什么这样判断?" | 单个决策的审计轨迹 | | `/pace-trace [CR] timeline` | "这个 CR 经历了什么?" | CR 全生命周期决策时间线 | | `/pace-status trace <name>` | "这个目标完成到哪了?" | 价值链完成度(OBJ→Epic→BR→PF→CR→OPP) | | `/pace-theory why` | "devpace 为什么这样设计?" | 系统方法论解释 |
输入
- `$ARGUMENTS`:可选——CR 名称/编号 + 决策类型
- 无参数时:展示当前/最近 CR 的最新决策轨迹
流程
执行路由表
固定加载
| 文件 | 说明 | |------|------| | `skills/pace-trace/trace-procedures-common.md` | 通用规则:§0 路由索引 · 输出格式框架 · confidence 表 · 降级处理 · 执行步骤 · 规则 |
按子命令加载
| 决策类型 | 额外加载文件 | 说明 | |---------|------------|------| | `gate1` / `gate2` / `gate3` | `skills/pace-trace/trace-procedures-gates.md` | Gate 类型输出要点 + 导航建议 | | `intent` / `change` / `risk` / `autonomy` | `skills/pace-trace/trace-procedures-analysis.md` | 分析类输出调整 + 导航建议 | | `timeline` | `skills/pace-trace/trace-procedures-timeline.md`(自包含,不加载 common) | CR 全生命周期决策时间线 | | `arch` | `skills/pace-trace/trace-procedures-arch.md`(自包含,不加载 common) | 架构决策记录(ADR)管理 | | 无指定 | 查 CR 事件表最新 checkpoint 类型,匹配上述路由 | 先确定类型再加载对应文件 |
**路由规则**:只读取路由表映射的 procedures 文件,不加载其他文件。timeline 自包含,不加载 common。
输出
格式化的决策轨迹(含判定结论、confidence 评估、评估过程、决策上下文、溯源信息、上下文导航)。
Read more
description: Use when user asks "why did devpace decide X", "追溯", "为什么这样做", "决策记录", "决策原因", "架构决策", "ADR", "技术选型", wants to see AI decision trail or manage architecture decisions, or says /pace-trace [CR] [gate/decision/arch]. NOT for concept explanations or methodology questions (use /pace-theory). NOT for metrics or retrospective analysis (use /pace-retro). allowed-tools: AskUserQuestion, Read, Write, Edit, Glob, Grep argument-hint: "[CR 名称或编号] [gate1|gate2|gate3|intent|change|risk|autonomy|timeline|arch]" model: haiku
/pace-trace — 决策轨迹查询
> 三层渐进透明的深入层。提供完整的 AI 决策轨迹,让用户可审计 Claude 的每个系统判断。
与相关功能的区别
| 功能 | 回答的问题 | 数据维度 | |------|-----------|---------| | `/pace-trace [CR] [type]` | "Claude 为什么这样判断?" | 单个决策的审计轨迹 | | `/pace-trace [CR] timeline` | "这个 CR 经历了什么?" | CR 全生命周期决策时间线 | | `/pace-status trace <name>` | "这个目标完成到哪了?" | 价值链完成度(OBJ→Epic→BR→PF→CR→OPP) | | `/pace-theory why` | "devpace 为什么这样设计?" | 系统方法论解释 |
输入
- `$ARGUMENTS`:可选——CR 名称/编号 + 决策类型
- 无参数时:展示当前/最近 CR 的最新决策轨迹
流程
执行路由表
固定加载
| 文件 | 说明 | |------|------| | `skills/pace-trace/trace-procedures-common.md` | 通用规则:§0 路由索引 · 输出格式框架 · confidence 表 · 降级处理 · 执行步骤 · 规则 |
按子命令加载
| 决策类型 | 额外加载文件 | 说明 | |---------|------------|------| | `gate1` / `gate2` / `gate3` | `skills/pace-trace/trace-procedures-gates.md` | Gate 类型输出要点 + 导航建议 | | `intent` / `change` / `risk` / `autonomy` | `skills/pace-trace/trace-procedures-analysis.md` | 分析类输出调整 + 导航建议 | | `timeline` | `skills/pace-trace/trace-procedures-timeline.md`(自包含,不加载 common) | CR 全生命周期决策时间线 | | `arch` | `skills/pace-trace/trace-procedures-arch.md`(自包含,不加载 common) | 架构决策记录(ADR)管理 | | 无指定 | 查 CR 事件表最新 checkpoint 类型,匹配上述路由 | 先确定类型再加载对应文件 |
**路由规则**:只读取路由表映射的 procedures 文件,不加载其他文件。timeline 自包含,不加载 common。
输出
格式化的决策轨迹(含判定结论、confidence 评估、评估过程、决策上下文、溯源信息、上下文导航)。
Give your Claude Code projects a steady development pace — requirements change, rhythm stays. A development harness for Claude Code — rules, schemas, gates, and feedback loops that keep AI-assisted development traceable and measurable.
Repo: arch-team/devpace
Other skills on devpace.
- /pace-biz
Use when user says "业务机会/OPP", "专题/Epic", "分解需求/decompose", "精炼/refine", "战略对齐/align", "业务全景/view", "需求发现/brainstorm/discover", "导入/import", "技术债务/infer", "pace-biz", or wants to plan requirements. NOT for /pace-dev, -change, -plan, -init.
Open skill - /pace-change
Use when user says "不做了", "搁置", "停掉", "加一个", "加需求", "追加", "改需求", "需求变了", "范围变了", "优先级调", "延后", "提前", "砍掉", "先做这个", "排到前面", "恢复", "捡回来", "pace-change", or wants to add, pause, resume, reprioritize, modify, undo, batch change, or query change history. NOT for code implementation
Open skill - /pace-dev
Use when user says "开始做", "帮我改", "实现", "修复", "继续推进", "编码", "写代码", "开发", "重构", "做个", "coding", "implement", "fix", "refactor", "build", /pace-dev, or explicitly requests to start/continue/resume coding work. "帮我改" only when target is code/UI/config. NOT for requirement changes
Open skill - /pace-feedback
Use when user reports issues, shares feedback, or receives production alerts — "用户反馈", "线上问题", "生产问题", "告警", "改进建议", "新需求", "体验问题", "功能请求", "线上bug", "运维", "事件", "incident", "故障", "P0", "P1", "严重故障", "postmortem", "事后复盘". NOT for code implementation or development (use
Open skill - /pace-guard
Use when user wants to assess risks before development, check current risk status, analyze risk trends, or says "风险/预检/预分析/guard/risk/隐患/安全检查". Also auto-invoked during advance mode intent checkpoint for L/XL CRs. NOT for /pace-dev (implementation), NOT for /pace-review (quality
Open skill - /pace-help
Use when user says "帮助", "help", "怎么用", "使用说明", "用法", "有哪些命令", "能做什么", "命令列表", "怎么入门", "quickstart", "FAQ", "常见问题", "pace-help", or wants help with how to use devpace commands and features. NOT for concept deep dive or methodology (use /pace-theory). NOT for next-step
Open skill

