Retrospective: [Sprint N / Milestone Name]【免费下载链接】Claude-Code-Game-StudiosTurn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.项目地址: https://gitcode.com/GitHub_Trending/cl/Claude-Code-Game-StudiosPeriod: [Start Date] -- [End Date] Generated: [Date]MetricsMetricPlannedActualDeltaTasks[X][Y][/- Z]Completion Rate--[Z%]--Story Points / Effort Days[X][Y][/- Z]Bugs Found--[N]--Bugs Fixed--[N]--Unplanned Tasks Added--[N]--Commits--[N]--- **Tasks**计划任务数来自 sprint 计划与完成数来自 git 与状态文件的对照 - **Completion Rate**实际完成率Phase 3 五分类的直接汇总 - **Story Points / Effort Days**若项目使用故事点或人日估算则填计划值、实际值及其差额 - **Bugs Found / Bugs Fixed**周期内发现与修复的缺陷数 - **Unplanned Tasks Added**中途插入的计划外任务——数值越大说明计划稳定性越差 - **Commits**git log 统计的实际提交数。 该表与 post-sprint hook 模板见第六节中的 Metrics 表字段基本一致Tasks Planned、Tasks Completed、Carryover from Previous、New Tasks Added、Bugs Found、Bugs Fixed可见项目把“计划 vs 完成 vs 缺陷 vs 计划外”视为回顾指标的最小公分母。 ### 5.2 Velocity Trend 速度趋势 markdown ### Velocity Trend | Sprint | Planned | Completed | Rate | |--------|---------|-----------|------| | [N-2] | [X] | [Y] | [Z%] | | [N-1] | [X] | [Y] | [Z%] | | [N] (current) | [X] | [Y] | [Z%] | **Trend**: [Increasing / Stable / Decreasing] [One sentence explaining the trend]至少列出最近三个 sprint 的计划量、完成量与完成率给出趋势判断并附一句解释。这份趋势数据会被 Phase 6 直接交给/sprint-plan用于下一轮排期——在 sprint-plan/SKILL.md 的 Phase 1 中明确写着“读取上一份 sprint 以理解 velocity 与 carryover”回顾与计划因此形成闭环。5.3 What Went Well / What Went Poorly### What Went Well - [Observation backed by specific data or examples] - [Another positive observation] - [Recognize specific contributions or decisions that paid off] ### What Went Poorly - [Specific issue with measurable impact -- e.g., Feature X took 5 days instead of estimated 2, blocking tasks Y and Z] - [Another issue with impact] - [Do not assign blame -- focus on systemic causes]两节共用一条核心纪律每一条观察都必须附带具体数据或实例严禁“沟通可以更好”这类空泛表述见 Guidelines 原文“Vague retrospectives produce vague improvements”。失败项要量化影响多花了几天、阻塞了哪些任务并且明确禁止归咎个人——只分析系统性成因。这与 .claude/docs/templates/post-mortem.md 中“What Went Poorly → Root cause → Impact → Prevention”的四段式结构共享同一理念追根因、不追责。5.4 Blockers Encountered 阻塞清单### Blockers Encountered | Blocker | Duration | Resolution | Prevention | |---------|----------|------------|------------| | [What blocked progress] | [How long] | [How it was resolved] | [How to prevent recurrence] |每行一个阻塞描述、持续时长、最终如何解决、以及防止复发的手段。post-sprint hook 的 Blockers 表Blocker / Frequency / Resolution Time / Prevention多了“频率”维度两者可以互相补充——频率字段特别适合识别“反复出现的同一类阻塞”。5.5 Estimation Accuracy 估算准确度### Estimation Accuracy | Task | Estimated | Actual | Variance | Likely Cause | |------|-----------|--------|----------|--------------| | [Most overestimated task] | [X] | [Y] | [Z] | [Why] | | [Most underestimated task] | [X] | [Y] | [-Z] | [Why] | **Overall estimation accuracy**: [X%] of tasks within /- 20% of estimate [Analysis: Are we consistently over- or under-estimating? For which types of tasks? What adjustment should we apply?]这一节同时给出总体精度在 ±20% 偏差内的任务占比与极端样本最被高估/低估的任务并要求给出系统性结论我们是否持续高估或低估哪种类型的任务偏差最大下一轮应如何调整估算方法——结论直接服务于/sprint-plan下一轮的能力预算。5.6 Carryover Analysis 结转分析### Carryover Analysis | Task | Original Sprint | Times Carried | Reason | Action | |------|----------------|---------------|--------|--------| | [Task that was not completed] | [Sprint N-X] | [N] | [Why] | [Complete / Descope / Redesign] |专门追踪“被反复结转的任务”记录它最初属于哪个 sprint、已被结转几次、原因并给出明确处置——完成 / 砍掉 / 重新设计。这一节与 sprint 计划中的 “Carryover from Previous Sprint” 表格Task | Reason | New Estimate形成前后对照计划时记录结转任务回顾时裁决其去留。5.7 技术债状态### Technical Debt Status - Current TODO count: [N] (previous: [N]) - Current FIXME count: [N] (previous: [N]) - Current HACK count: [N] (previous: [N]) - Trend: [Growing / Stable / Shrinking] - [Note any areas of concern]输出三类质量标记的当前计数与历史计数判定趋势并标注关注区域。数据取自 Phase 3 的代码扫描是判断“技术债是否失控”的量化锚点。5.8 历史行动项跟进### Previous Action Items Follow-Up | Action Item (from Sprint N-1) | Status | Notes | |-------------------------------|--------|-------| | [Previous action] | [Done / In Progress / Not Started] | [Context] |对照上一份回顾逐项核验行动项状态。技能 Guidelines 特别强调“Recurring unaddressed items are a process smell”——若同一行动项反复出现在各期回顾中却从未落实这本身就是要进入下一轮 Action Items 的首要问题。5.9 下一轮行动项### Action Items for Next Iteration | # | Action | Owner | Priority | Deadline | |---|--------|-------|----------|----------| | 1 | [Specific, measurable action] | [Who] | [High/Med/Low] | [When] | | 2 | [Another action] | [Who] | [Priority] | [When] |每条行动项必须有负责人Owner、优先级与截止时间且总数控制在 3–5 条见 Guidelines“Limit action items to 3-5. More than that dilutes focus”。这是整份回顾中“可执行性”的最终落点——其余章节都是诊断只有这里产生处方。5.10 流程改进与总结### Process Improvements - [Specific change to how we work, with expected benefit] - [Another improvement -- keep it to 2-3 actionable items, not a wish list] ### Summary [2-3 sentence overall assessment: Was this a good sprint/milestone? What is the single most important thing to change going forward?]Process Improvements 限定 2–3 条具体的做事方式改变并说明预期收益避免变成愿望清单Summary 用 2–3 句话给出整体评价并点名“接下来最需要改变的一件事”。六、Phase 5保存回顾与 Verdict生成完成后技能先向用户口头呈现核心发现完成率、速度趋势、最大阻塞、最重要的行动项然后询问May I write this toproduction/sprints/sprint-[N]-retrospective.md?或里程碑对应路径用户同意 → 写入文件必要时自动创建目录判定COMPLETE——回顾已保存用户拒绝 → 停止判定BLOCKED——用户拒绝写入。注意 verdict 机制贯穿整个项目sprint 计划的成功保存是COMPLETE、失败是BLOCKED见 sprint-plan/SKILL.md 的 Phase 4里程碑回顾同理见 milestone-review/SKILL.md 的 Phase 4。Agent 从不擅自落盘写文件前永远征得用户同意——这是本项目协作协议Collaborative Protocol的铁律。七、Phase 6后续步骤与使用准则回顾产出流向哪里保存之后技能给出两条明确的下游动作若刚完成 sprint 回顾 → 运行/sprint-plan把行动项与 velocity 数据吸收进下一轮 sprint 计划若刚完成里程碑回顾 → 运行/gate-check正式评估进入下一开发阶段的就绪度。这两条链路都有仓库级实现支撑sprint-plan/SKILL.md 的 Phase 1 明确“读取上一份 sprint 以理解 velocity 与 carryover”其任务表Must Have / Should Have / Nice to Have、Carryover 表与风险表可直接承接回顾结论gate-check/SKILL.md 定义了从 Concept 到 Release 的七个阶段及每道门的制品清单里程碑回顾判定“目标是否达成”后正好用阶段门禁决定是否可以前进。Guidelines 使用准则原文语义- Be honest and specific. Vague retrospectives (communication could be better) produce vague improvements. Use data and examples. - Focus on systemic issues, not individual blame. - Limit action items to 3-5. More than that dilutes focus. - Every action item must have an owner and a deadline. - Check whether previous action items were completed. Recurring unaddressed items are a process smell. - If this is a milestone retrospective, also evaluate whether the milestone goals were achieved and what that means for the overall project timeline.六条准则可归纳为三组纪律诚实具体数据与实例优先、系统归因不追责个人、行动收敛行动项限量、必有负责人与截止时间、必须回溯历史行动项。里程碑回顾的额外职责是评估目标达成度及其对整体时间线的意义——这正是 /milestone-review 中 Go/No-Go 评估的前置输入。八、仓库协同post-sprint hook 与 producer 角色post-sprint-retrospective 工作流 Hook仓库在 .claude/docs/hooks-reference/post-sprint-retrospective.md 定义了一个工作流 Hook非 git Hook作为/retrospective的自动化入口。它不是通过 git 事件触发而是由用户或 producer 代理显式调用producer Generate sprint retrospective for Sprint [N]【免费下载链接】Claude-Code-Game-StudiosTurn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.项目地址: https://gitcode.com/GitHub_Trending/cl/Claude-Code-Game-Studios创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考