pm-retro

Preamble

echo "📝 迭代复盘工具已启动"

# 检查迭代计划
if [ -f "docs/03-增长迭代/迭代计划.md" ]; then
  echo "✅ 迭代计划 - 已找到"
else
  echo "⏳ 迭代计划 - 未找到"
fi

# 检查迭代完成情况
if [ -f "docs/03-增长迭代/数据报告与用户反馈.md" ]; then
  echo "✅ 数据报告 - 已找到"
else
  echo "⏳ 数据报告 - 未找到"
fi

Write-time boundaries

When only discussing, do not write files; during startup checks, do not create directories. When the user asks to save artifacts, first confirm the current product, the exact output path, and the overwrite/append choice for existing files; after confirmation,only immediately before the actual Write,create the parent directory of that output file. For full installation, see skills/_shared/artifact-policy.md, and installing this skill alone also follows this paragraph.

Cross-Agent interaction rules

When the process requires interaction with the user:

If the current environment supports AskUserQuestion, use AskUserQuestion (best experience).
If the current environment does not support AskUserQuestion, you must ask the same question via a normal chat message.
Ask only one question at a time.
After asking, you must stop the current turn and wait for the user's answer (STOP and WAIT).
Do not generate documents or write to docs before the user answers.
Existing docs files cannot replace the user's answer in this turn.

Execution flow

Step 1: Retrospective preparation

Use AskUserQuestion to ask:

🎯 Iteration retrospective preparation

Iteration information：

Iteration number: Iteration X

Iteration period: YYYY-MM-DD to YYYY-MM-DD

Participating roles: Product, Development, Design, QA, Operations (multiple selections allowed)

Iteration goal review：

What were the core goals originally set for this iteration?

A) Goal 1: {description} B) Goal 2: {description} C) Goal 3: {description} D) No clear goals

Record to variable ITERATION_INFO

Step 2: Goal achievement assessment

📊 Goal achievement assessment

Goal 1: {goal description}

Completion status: {X}%

Achieved: ✅/🟡/❌

Explanation: {reason for completion/non-completion}

Goal 2: {goal description}

Completion status: {X}%

Achieved: ✅/🟡/❌

Explanation: {reason for completion/non-completion}

Overall completion rate：{X}%

Requirement completion statistics：

Priority Planned Completed Not completed Completion rate

P0 X X X X%

P1 X X X X%

P2 X X X X%

Is the above data accurate?

Priority	Planned	Completed	Not completed	Completion rate
P0	X	X	X	X%
P1	X	X	X	X%
P2	X	X	X	X%

Step 3: Collect team feedback (KPT)

Use AskUserQuestion to collect KPT (Keep-Problem-Try):

📝 Team feedback collection (KPT model)

Keep - What went well (worth continuing)

Examples:

Standardized code review process, reduced bug rate

High efficiency in daily stand-up meetings

Timely design reviews

Please list what went well (one per line, enter "done" to finish):

Collected list KEEP_ITEMS

Problem - Issues that need improvement

Examples:

Frequent requirement changes, affecting development pace

Insufficient testing time, more online bugs

Cross-department communication not timely

Please list the issues that need improvement (one per line, enter "done" to finish):

Collected list PROBLEM_ITEMS

Try - Action plans to try

For the above issues, try the following actions:

Example:

Establish a requirement change approval mechanism

Reserve 20% testing buffer time

Establish a cross-department communication group and sync progress daily

Please list the actions to try (one per line, enter "done" to finish):

Collected list TRY_ITEMS

Step 4: Root cause analysis

Conduct root cause analysis on the TOP issue (the one with the most team feedback):

🔍 Root cause analysis

TOP issue: {issue description}

5 Whys analysis：
问题：{问题}
Why 1：{直接原因}
Why 2：{深层原因}
Why 3：{根本原因}
Why 4：{系统性原因}
Why 5：{根因结论}
Root cause: {root cause conclusion}

Solution：

{short-term measure} - execute immediately

{long-term measure} - continuous improvement

{preventive measure} - prevent recurrence

Step 5: Formulate improvement measures

🎯 Improvement measures and action items

Action item Owner Deadline Priority

{action item 1} {owner} {date} P0

{action item 2} {owner} {date} P1

{action item 3} {owner} {date} P1

Improvement goals for the next iteration：

Increase the P0 completion rate from {X}% to {Y}%

Resolve the TOP issue: {issue} to {standard}

Action item	Owner	Deadline	Priority
{action item 1}	{owner}	{date}	P0
{action item 2}	{owner}	{date}	P1
{action item 3}	{owner}	{date}	P1

Step 6: Output the iteration retrospective report

Use the Write tool to create docs/03-增长迭代/迭代复盘/复盘-v{版本号}.md：

# 迭代复盘报告 v{版本号}

## 一、迭代概览

- **迭代编号**：第X次迭代
- **周期**：YYYY-MM-DD 至 YYYY-MM-DD
- **参与角色**：{角色列表}

## 二、目标达成

| 目标 | 完成率 | 状态 | 说明 |
|------|--------|------|------|
| {目标1} | X% | ✅/🟡/❌ | {说明} |
| {目标2} | X% | ✅/🟡/❌ | {说明} |
| **整体** | **X%** | | |

## 三、Keep - 做得好的

1. {内容}
2. {内容}
3. {内容}

## 四、Problem - 需要改进的

1. {内容}（根因：{根因}）
2. {内容}
3. {内容}

## 五、Try - 尝试的行动

| 行动项 | 负责人 | 截止时间 | 优先级 |
|--------|--------|---------|--------|
| {行动} | {负责人} | {日期} | P0 |
| {行动} | {负责人} | {日期} | P1 |

## 六、数据指标

| 指标 | 迭代前 | 迭代后 | 变化 |
|------|--------|--------|------|
| {指标} | X | Y | ±Z% |

## 七、下个迭代改进目标

1. {目标1}
2. {目标2}

---

**文档状态**: 迭代复盘完成
**生成时间**: {时间戳}
**生成工具**: super-pm

Step 7: Completion prompt

✅ Iteration retrospective complete!

📄 Generated:docs/03-增长迭代/迭代复盘/复盘-v{版本号}.md

🎯 Suggested next steps:

A) Run /pm-iteration - formulate the next iteration plan B) Run /pm-roadmap - update the product roadmap C) If the improvement measures involve risks, run /pm-risk

Fallback mechanism

Scenario: No prior iteration plan

If there is no iteration plan document, go directly to the KPT retrospective process.

Notes

Safe atmosphere: A retrospective is not about assigning blame; focus on process improvement rather than individuals
Let data speak: Goal achievement should be supported by data, avoiding subjective judgment
Action-driven: Every issue must have a corresponding action item
Tracking loop closure: At the next retrospective, first review the completion status of this retrospective's action items

Output quality comparison

✅ Good example：

- 有数据引用：「根据 Q4 数据，留存率从 35% 降至 28%」
- 有验证来源：「数据来源：Google Analytics, 2025-12-01」
- 有明确建议：「建议将新手引导步骤从 5 步减少至 3 步」

❌ Bad example：

- 模糊结论：「数据表明留存率有所下降」
- 无来源：「根据经验，这个功能很重要」
- 没有行动建议：「留存是个问题」

Common pitfalls / Red Flags — STOP

Stop immediately and backtrack if any of the following occur:

Pitfall	Correct approach
Use "should", "probably", "seems" to draw conclusions	Must be based on actual data and verification
Claim completion without running checks	Verify first, then state
Skip key steps due to time pressure	No exceptions; the tighter the time, the stricter you must be
The thought that "this time it should be fine"	Re-verify every time

Output Quality Check / Verification Checklist

Prerequisite dependencies are satisfied (input documents/data have been collected)
All core steps have been executed
Output documents have been generated to the docs/ directory
Every judgment is supported by data/evidence
2-3 follow-up skills have been recommended

⚠️ If any item does not pass → complete it before marking as done.