Initial commit

2025-11-30 08:38:26 +08:00
commit 41d9f6b189
304 changed files with 98322 additions and 0 deletions
--- a/skills/reviews-retros-reflection/SKILL.md
+++ b/skills/reviews-retros-reflection/SKILL.md
@@ -0,0 +1,196 @@
+---
+name: reviews-retros-reflection
+description: Use when conducting sprint retrospectives, project post-mortems, weekly reviews, quarterly reflections, after-action reviews (AARs), team health checks, process improvement sessions, celebrating wins while learning from misses, establishing continuous improvement habits, or when user mentions "retro", "retrospective", "what went well", "lessons learned", "review meeting", "reflection", or "how can we improve".
+---
+
+# Reviews, Retros & Reflection
+
+## Table of Contents
+1. [Purpose](#purpose)
+2. [When to Use](#when-to-use)
+3. [What Is It](#what-is-it)
+4. [Workflow](#workflow)
+5. [Retrospective Formats](#retrospective-formats)
+6. [Common Patterns](#common-patterns)
+7. [Guardrails](#guardrails)
+8. [Quick Reference](#quick-reference)
+
+## Purpose
+
+Reviews, Retros & Reflection helps teams and individuals systematically learn from experience through structured reflection, root cause analysis, and actionable improvement planning. It creates psychological safety for honest feedback while driving measurable progress.
+
+## When to Use
+
+**Invoke this skill when you need to:**
+- Conduct sprint/iteration retrospectives (Agile, Scrum teams)
+- Run project post-mortems or after-action reviews
+- Facilitate weekly/monthly team reviews
+- Reflect on quarterly or annual performance
+- Process significant events (launch, incident, milestone)
+- Improve team dynamics and collaboration
+- Establish continuous learning culture
+- Celebrate successes while extracting lessons from failures
+- Generate actionable improvement items
+- Build team alignment on priorities
+
+**User phrases that trigger this skill:**
+- "Let's do a retro"
+- "What went well/wrong?"
+- "How can we improve?"
+- "Lessons learned from..."
+- "Sprint retrospective"
+- "After-action review"
+- "Team reflection session"
+- "Review our progress"
+
+## What Is It
+
+A structured reflection process that:
+1. **Gathers data** (what happened during period)
+2. **Generates insights** (why it happened, patterns)
+3. **Decides actions** (what to change, keep, start, stop)
+4. **Tracks progress** (follow-up on previous actions)
+5. **Celebrates wins** (recognize successes, build morale)
+
+**Quick example (Start/Stop/Continue format):**
+- **Start**: Daily 15-min standup (async in Slack), mob programming for complex features
+- **Stop**: Last-minute scope changes, weekend deployments
+- **Continue**: Pairing on new features, celebrating small wins in team channel
+- **Actions**: (1) Create standup bot template by Friday, (2) Add "no scope changes <3 days before sprint end" to team charter
+
+## Workflow
+
+Copy this checklist and track your progress:
+
+```
+Retrospective Progress:
+- [ ] Step 1: Set the stage (context, psychological safety)
+- [ ] Step 2: Gather data (what happened)
+- [ ] Step 3: Generate insights (why it happened)
+- [ ] Step 4: Decide actions (what to change)
+- [ ] Step 5: Close and follow up (commit, track)
+```
+
+**Step 1: Set the stage**
+
+Define period/scope, review previous action items, establish psychological safety (Prime Directive: "everyone did best job given knowledge/skills/resources/context"). For quick reviews → Use [resources/template.md](resources/template.md). For complex team retros → Study [resources/methodology.md](resources/methodology.md).
+
+**Step 2: Gather data**
+
+Collect facts about period: metrics (velocity, bugs, incidents), events (launches, blockers, decisions), sentiment (team energy, morale). See [Retrospective Formats](#retrospective-formats) for collection methods.
+
+**Step 3: Generate insights**
+
+Identify patterns, root causes, surprises. Ask "why?" to move from symptoms to causes. Use [resources/methodology.md](resources/methodology.md) for root cause techniques (5 Whys, fishbone diagrams, timeline analysis).
+
+**Step 4: Decide actions**
+
+Vote on most impactful improvements (dot voting, SMART criteria). Define 1-3 SMART actions (Specific, Measurable, Assigned owner, Realistic, Time-bound). See [Common Patterns](#common-patterns) for action quality criteria.
+
+**Step 5: Close and follow up**
+
+Commit to actions, schedule check-in, thank participants. Track action completion rate (target: >80% completion before next retro). Self-check using `resources/evaluators/rubric_reviews_retros_reflection.json` before closing. Minimum standard: Average score ≥ 3.5.
+
+## Retrospective Formats
+
+**Start/Stop/Continue (Simple, Balanced):**
+- **Start**: What should we begin doing?
+- **Stop**: What should we stop doing?
+- **Continue**: What's working well, keep doing?
+- **When**: General purpose, new teams, tight on time (30 min)
+
+**Mad/Sad/Glad (Emotion-Focused):**
+- **Mad**: What frustrated or angered you?
+- **Sad**: What disappointed you?
+- **Glad**: What made you happy?
+- **When**: Processing difficult period, improving morale, surfacing hidden issues
+
+**4Ls (Comprehensive):**
+- **Loved**: What did we love?
+- **Learned**: What did we learn?
+- **Lacked**: What did we lack?
+- **Longed for**: What did we long for?
+- **When**: End of major project, quarterly reviews, want depth
+
+**Sailboat/Speedboat (Metaphor-Based):**
+- **Wind** (helping): What's propelling us forward?
+- **Anchor** (hindering): What's slowing us down?
+- **Rocks** (risks): What dangers lie ahead?
+- **Island** (goal): Where are we going?
+- **When**: Strategic planning, visualizing progress, cross-functional teams
+
+**Timeline (Chronological):**
+- Plot events on timeline, mark highs/lows, identify patterns
+- **When**: Long period (quarter), complex project, need shared understanding
+
+## Common Patterns
+
+**Pattern 1: Sprint Retrospective (Agile)**
+- **Frequency**: Every 1-2 weeks
+- **Duration**: 45-90 min (shorter sprints = shorter retros)
+- **Focus**: Process improvements, team dynamics, technical practices
+- **Format**: Start/Stop/Continue or Mad/Sad/Glad
+- **Actions**: 1-3 process improvements, 1 technical improvement
+
+**Pattern 2: Project Post-Mortem**
+- **Frequency**: End of project/phase
+- **Duration**: 90-120 min
+- **Focus**: What to repeat, what to avoid, systemic issues
+- **Format**: 4Ls or Timeline
+- **Actions**: Documentation updates, playbook changes, skill gaps to address
+
+**Pattern 3: Weekly Team Review**
+- **Frequency**: Weekly (Fridays common)
+- **Duration**: 15-30 min
+- **Focus**: Wins, blockers, priorities for next week
+- **Format**: Custom (Wins/Blockers/Priorities)
+- **Actions**: Blocker removal, celebrate wins, align on top 3 priorities
+
+**Pattern 4: Incident Retrospective**
+- **Frequency**: After major incidents
+- **Duration**: 60 min
+- **Focus**: Blameless analysis, system improvements
+- **Format**: Timeline + 5 Whys
+- **Actions**: Incident response improvements, monitoring/alerting, prevention
+
+## Guardrails
+
+**Psychological safety:**
+- **Prime Directive**: "Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, and the situation at hand."
+- Focus on processes/systems, not people
+- No blame, no judgment, no defensiveness
+- Encourage dissenting opinions
+- What's said in retro stays in retro (unless actionable for others)
+
+**Quality standards:**
+- **Action items are SMART**: Specific, Measurable, Assigned, Realistic, Time-bound
+- **Track completion**: >80% action completion before next retro (if <80%, too many actions or not committed)
+- **Rotate facilitator**: Different person each time, prevents bias
+- **Time-box discussions**: Don't get stuck, move on, revisit if time
+- **Vote for focus**: Use dot voting to prioritize discussion topics
+
+**Common pitfalls to avoid:**
+- **Too many actions**: >5 actions = none get done. Focus on 1-3 high-impact items.
+- **Vague actions**: "Improve communication" → ❌. "Daily 15-min standup in #eng-sync by Monday" → ✓
+- **No follow-up**: Actions from last retro ignored → trust erodes, retros become theater
+- **Blame culture**: Pointing fingers → people stop being honest
+- **Same issues every retro**: If no progress on recurring issues, escalate to leadership
+
+## Quick Reference
+
+**Resources:**
+- **Quick retro formats**: [resources/template.md](resources/template.md)
+- **Facilitation techniques**: [resources/methodology.md](resources/methodology.md)
+- **Quality rubric**: `resources/evaluators/rubric_reviews_retros_reflection.json`
+
+**5-Stage Process**: Set Stage → Gather Data → Generate Insights → Decide Actions → Close
+
+**Top Formats**:
+- **Start/Stop/Continue**: General purpose, 30 min
+- **Mad/Sad/Glad**: Emotion processing, 45 min
+- **4Ls**: Deep reflection, 60 min
+- **Sailboat**: Visual/strategic, 60 min
+
+**Action Quality**: SMART criteria + <5 total + >80% completion rate
+
+**Psychological Safety**: Prime Directive + Blameless + Confidential
--- a/skills/reviews-retros-reflection/resources/evaluators/rubric_reviews_retros_reflection.json
+++ b/skills/reviews-retros-reflection/resources/evaluators/rubric_reviews_retros_reflection.json
@@ -0,0 +1,171 @@
+{
+  "name": "Reviews, Retros & Reflection Evaluator",
+  "description": "Evaluates retrospectives for psychological safety, data-driven insights, actionable outcomes, and continuous improvement focus",
+  "criteria": [
+    {
+      "name": "Psychological Safety and Blamelessness",
+      "weight": 1.4,
+      "scale": {
+        "1": "Blame culture evident (finger pointing, defensiveness, personal attacks)",
+        "2": "Some blame language, people-focused criticism, not fully safe to speak",
+        "3": "Mostly blameless, some slips into people-focus, Prime Directive mentioned but not enforced",
+        "4": "Consistently blameless, systems-focused, Prime Directive established, encourages honest feedback",
+        "5": "Exemplary: Prime Directive explicitly stated and reinforced, completely blameless language, focuses on systems/processes not people, psychological safety actively built (confidentiality, no judgment), dissenting opinions encouraged and valued"
+      }
+    },
+    {
+      "name": "Data Gathering Quality",
+      "weight": 1.3,
+      "scale": {
+        "1": "Vague recollections only ('things were bad'), no concrete data or examples",
+        "2": "Some anecdotes but mostly opinions, missing metrics or specific events",
+        "3": "Mix of data (metrics, events) and sentiment, basic coverage of period",
+        "4": "Comprehensive data: metrics (velocity, bugs, incidents), events (launches, decisions), sentiment (team energy), specific examples",
+        "5": "Exemplary: Rich data from multiple sources (metrics dashboard, timeline of events, team surveys, customer feedback), balanced quantitative and qualitative, patterns identified across data, surprises and wins highlighted"
+      }
+    },
+    {
+      "name": "Insight Generation and Root Cause Analysis",
+      "weight": 1.4,
+      "scale": {
+        "1": "No insights, just list of complaints or surface observations",
+        "2": "Basic observations but no depth, symptoms identified but not root causes",
+        "3": "Some root cause analysis (one level of 'why'), patterns noted, themes identified",
+        "4": "Deep root cause analysis (5 Whys, fishbone, timeline analysis), clear themes with supporting evidence, distinguishes symptoms from causes",
+        "5": "Exemplary: Rigorous root cause techniques applied, patterns across multiple retros identified, systemic issues surfaced (not just immediate causes), surprising insights discovered, connections made between seemingly unrelated items, both process and cultural factors explored"
+      }
+    },
+    {
+      "name": "Action Quality (SMART Criteria)",
+      "weight": 1.5,
+      "scale": {
+        "1": "Vague actions ('improve communication', 'do better'), no specificity or ownership",
+        "2": "Actions stated but missing SMART elements (no owner, deadline, or measurability)",
+        "3": "Actions are mostly SMART but some elements weak (vague success criteria, unrealistic scope)",
+        "4": "All actions are SMART (Specific, Measurable, Assigned, Realistic, Time-bound), clear and achievable",
+        "5": "Exemplary: Actions exceed SMART criteria (addresses root causes not symptoms, team has authority to implement, concrete first steps defined, success criteria binary or numeric, owners present and committed, 1-3 high-impact actions not overwhelming list)"
+      }
+    },
+    {
+      "name": "Participation and Inclusivity",
+      "weight": 1.2,
+      "scale": {
+        "1": "One or two voices dominate, most team silent or absent",
+        "2": "Unbalanced participation, some voices not heard, no facilitation to balance",
+        "3": "Most participate but some imbalance, facilitator makes some effort to include all",
+        "4": "Balanced participation, facilitator actively ensures all voices heard, techniques used (silent brainstorming, round-robin)",
+        "5": "Exemplary: All voices heard and valued, silent/async methods used to balance introverts/extroverts, dominant voices managed tactfully, psychological safety enables honest participation from all, diverse perspectives explicitly sought"
+      }
+    },
+    {
+      "name": "Follow-Up and Accountability",
+      "weight": 1.3,
+      "scale": {
+        "1": "No review of previous actions, no tracking, actions forgotten after retro",
+        "2": "Previous actions mentioned but not tracked, no completion metrics, unclear accountability",
+        "3": "Previous actions reviewed with some status updates, basic tracking, completion rate noted",
+        "4": "Previous actions systematically reviewed (done/in-progress/blocked), completion rate calculated (target >80%), owners accountable, follow-up scheduled",
+        "5": "Exemplary: Action completion tracked across retros (dashboard or log), completion rate >80% achieved, incomplete actions discussed (why? pivot or continue?), follow-up meeting scheduled, retro actions linked to work tracking system, team celebrates completed actions, retro-to-retro learning visible"
+      }
+    },
+    {
+      "name": "Format Appropriateness",
+      "weight": 1.1,
+      "scale": {
+        "1": "Format clearly mismatched (Timeline for 1-week sprint, or no structure at all)",
+        "2": "Generic format used without considering context (always same format regardless of need)",
+        "3": "Format somewhat appropriate but could be better matched to situation",
+        "4": "Format well-matched to context (period length, team needs, goals)",
+        "5": "Exemplary: Format deliberately chosen with rationale (Start/Stop/Continue for action focus, Mad/Sad/Glad for morale, 4Ls for depth, Timeline for complex period), adapted to team preferences and needs, variety across retros to prevent staleness"
+      }
+    },
+    {
+      "name": "Continuous Improvement Focus",
+      "weight": 1.2,
+      "scale": {
+        "1": "Retro feels like venting session with no improvement focus, no actions or same actions every time",
+        "2": "Some improvement intent but actions vague or not followed through, same issues recurring",
+        "3": "Clear improvement focus, some progress on issues, a few recurring themes addressed",
+        "4": "Strong improvement culture, actions tracked and completed, progress visible over time, recurring issues escalated",
+        "5": "Exemplary: Measurable improvement over time (metrics, team health, action completion >80%), retro process itself improves (ROTI scores, format experiments), learning documented and shared, recurring issues systematically addressed or escalated, celebration of progress and wins, team proactively requests retros"
+      }
+    }
+  ],
+  "guidance": {
+    "by_retro_type": {
+      "sprint_retro": {
+        "recommended_format": "Start/Stop/Continue or Mad/Sad/Glad (30-90 min)",
+        "key_focus": "Process improvements, team dynamics, technical practices",
+        "action_targets": "1-3 process improvements, 1 technical improvement",
+        "red_flags": ["Same issues >3 sprints without escalation", "No technical debt actions", ">5 actions (too many)"]
+      },
+      "project_postmortem": {
+        "recommended_format": "4Ls or Timeline (90-120 min)",
+        "key_focus": "Systemic learnings, documentation updates, playbook changes",
+        "action_targets": "Update documentation, refine playbooks, address skill gaps",
+        "red_flags": ["Blame culture (pointing fingers)", "No documentation of learnings", "Missing stakeholder perspectives"]
+      },
+      "weekly_review": {
+        "recommended_format": "Wins/Blockers/Priorities (15-30 min)",
+        "key_focus": "Celebrate wins, remove blockers, align on priorities",
+        "action_targets": "Blocker removal, top 3 priorities clear",
+        "red_flags": ["Skipped regularly (lose habit)", "All blockers, no wins (morale issue)", "Priorities change weekly (chaos)"]
+      },
+      "incident_retro": {
+        "recommended_format": "Timeline + 5 Whys (60 min)",
+        "key_focus": "Blameless analysis, prevention, monitoring improvements",
+        "action_targets": "Incident response process, monitoring/alerting, prevention measures",
+        "red_flags": ["Blame individuals (not systems)", "Stopped at symptom (didn't reach root cause)", "No follow-up on actions (incidents recur)"]
+      }
+    }
+  },
+  "common_failure_modes": {
+    "too_many_actions": {
+      "symptom": ">5 actions generated, <50% completion rate",
+      "root_cause": "Want to fix everything, don't prioritize, unrealistic about capacity",
+      "fix": "Dot vote for top 1-3 high-impact actions, defer others to future retros, track completion rate and adjust"
+    },
+    "vague_actions": {
+      "symptom": "Actions like 'improve communication', 'do better testing' without specifics",
+      "root_cause": "Didn't apply SMART criteria, rushed action planning, facilitator didn't push for specificity",
+      "fix": "Use SMART template explicitly, facilitator role-plays action completion to test specificity ('how exactly would you do this?')"
+    },
+    "blame_culture": {
+      "symptom": "Finger pointing, defensiveness, people stop speaking honestly, same safe topics every retro",
+      "root_cause": "Prime Directive not established, people-focused language, leadership doesn't model blamelessness",
+      "fix": "State Prime Directive explicitly every retro, shift language from 'you' to 'the system', facilitator redirects blame to process, leadership training on blameless culture"
+    },
+    "no_follow_up": {
+      "symptom": "Actions from previous retro ignored, <50% completion, trust erodes, retros feel like theater",
+      "root_cause": "No tracking system, no accountability, actions not integrated into work flow, too ambitious actions",
+      "fix": "Create action tracking dashboard (visible to team), review at start of every retro, integrate actions into sprint/work planning, reduce action count to achievable 1-3"
+    },
+    "same_issues_recurring": {
+      "symptom": "Same problem appears >3 retros with no progress",
+      "root_cause": "Team lacks authority to fix (systemic issue beyond team), actions don't address root cause, leadership not engaged",
+      "fix": "Escalate to leadership with data ('issue X mentioned in last 4 retros, team can't solve alone'), use root cause analysis to go deeper, request organizational support or decision"
+    },
+    "low_participation": {
+      "symptom": "<70% attendance, few people contribute, dominated by 1-2 voices",
+      "root_cause": "Retro not valued (no action follow-through), not psychologically safe, format is stale, poor facilitation",
+      "fix": "Improve action follow-through (builds trust), rotate facilitator, experiment with formats, use silent/async techniques to include all, one-on-ones to understand barriers"
+    }
+  },
+  "excellence_indicators": [
+    "Prime Directive stated explicitly and enforced throughout",
+    "Blameless language consistently used (systems-focused, not people-focused)",
+    "Comprehensive data gathered: metrics + events + sentiment",
+    "Root cause analysis conducted (5 Whys, fishbone, timeline)",
+    "Patterns and themes identified across observations",
+    "1-3 SMART actions generated (not overwhelming)",
+    "Actions address root causes, not just symptoms",
+    "All team members participated (balanced facilitation)",
+    "Previous actions reviewed with >80% completion rate",
+    "Follow-up meeting scheduled for action tracking",
+    "Format appropriate to context and team needs",
+    "Wins celebrated and recognized publicly",
+    "Continuous improvement focus evident (measurable progress over time)",
+    "ROTI score >80% thumbs up (retro itself is valuable)",
+    "Learning documented and shared beyond team when relevant"
+  ]
+}
--- a/skills/reviews-retros-reflection/resources/methodology.md
+++ b/skills/reviews-retros-reflection/resources/methodology.md
@@ -0,0 +1,405 @@
+# Reviews, Retros & Reflection: Advanced Methodologies
+
+## Table of Contents
+1. [Advanced Retrospective Formats](#1-advanced-retrospective-formats)
+2. [Facilitation Techniques](#2-facilitation-techniques)
+3. [Root Cause Analysis Methods](#3-root-cause-analysis-methods)
+4. [Psychological Safety Building](#4-psychological-safety-building)
+5. [Action Tracking and Metrics](#5-action-tracking-and-metrics)
+6. [Remote and Async Retrospectives](#6-remote-and-async-retrospectives)
+
+## 1. Advanced Retrospective Formats
+
+### Lean Coffee (Self-Organizing Agenda)
+
+**Setup**:
+- Participants write topics on cards
+- Group similar topics
+- Dot vote to prioritize
+- Discuss top topics time-boxed (5-7 min each)
+- Thumbs up/down/sideways to continue or move on
+
+**When**: Diverse team needs, unclear what to discuss, want emergent agenda
+
+**Pros**: Democratic, surfaces unexpected issues, adapts to group needs
+**Cons**: Can be unfocused, requires strong time-boxing
+
+### Perfection Game (Aspirational)
+
+**Process**:
+1. Rate period 1-10 (10 = perfect)
+2. "What did you like about this period?"
+3. "What would make it a 10?"
+4. Convert "make it 10" suggestions into actions
+
+**When**: Positive framing needed, team demoralized, focus on future not past
+
+**Pros**: Forward-looking, avoids negativity spiral, actionable
+**Cons**: Can avoid real problems if not pushed for honesty
+
+### Return on Time Invested (ROTI)
+
+**Process**:
+- Each participant rates retro value: thumb up (worth time), sideways (neutral), down (waste of time)
+- Anonymous or public depending on culture
+- Discuss: What made it valuable/not? How to improve next time?
+
+**When**: End of every retro to improve retro itself
+
+**Metrics to track**: % thumbs up over time (target: >80%)
+
+### Starfish (Five Categories)
+
+**Categories**:
+- **Keep Doing**: Works well, maintain
+- **Less Of**: Doing too much, scale back
+- **More Of**: Doing some, do more
+- **Stop Doing**: Not working, eliminate
+- **Start Doing**: Not doing, should begin
+
+**When**: More nuance than Start/Stop/Continue, want gradation
+
+**Pros**: Spectrum of actions, acknowledges partial successes
+**Cons**: More complex, can overwhelm with options
+
+### Speedboat with Crew (Team Dynamics Focus)
+
+**Extended metaphor**:
+- **Captain** (leadership): What's steering us?
+- **Crew** (team): What's propelling us?
+- **Wind** (external help): What's helping externally?
+- **Anchor** (drag): What's slowing us?
+- **Rocks** (risks): What dangers ahead?
+- **Shore** (safety): What's our safety net?
+- **Island** (goal): Where are we going?
+
+**When**: Complex team dynamics, cross-functional alignment, strategic context needed
+
+### Kaleidoscope (Multiple Perspectives)
+
+**Process**:
+- Divide team into small groups (3-4 people)
+- Each group discusses period from different lens:
+  - Customer perspective
+  - Business perspective
+  - Technical perspective
+  - Team health perspective
+- Groups share findings
+- Synthesize cross-perspective insights
+
+**When**: Cross-functional team, need multiple viewpoints, large retro (>10 people)
+
+## 2. Facilitation Techniques
+
+### Balancing Participation
+
+**Silent participants**:
+- Call on directly with specific question
+- Use round-robin (everyone speaks once before anyone twice)
+- Silent brainstorming before verbal discussion
+- Anonymous input (digital tools, sticky notes)
+
+**Dominant participants**:
+- "Let's hear from folks who haven't spoken yet"
+- Time limits per person
+- Parking lot for off-topic deep dives
+- Private conversation outside retro if pattern persists
+
+**Conflict emergence**:
+- Acknowledge tension, don't dismiss
+- Reframe as system/process issue, not personal
+- Focus on data/facts, not judgments
+- Table if too heated, address offline with parties
+
+### Time-Boxing Discipline
+
+**Visible timer**: Project countdown timer, everyone sees time remaining
+
+**Gentle warnings**: "3 minutes left on this topic"
+
+**Hard stops**: Move on even mid-discussion (capture in parking lot if needed)
+
+**Flex time**: Reserve 10-15 min at end for parking lot topics if time permits
+
+### Clustering and Affinity Mapping
+
+**Process**:
+1. Gather all items (sticky notes, digital cards)
+2. Read aloud without discussion
+3. Ask: "Which items are about the same thing?"
+4. Physically group similar items
+5. Name each cluster (theme)
+6. Dot vote on clusters, not individual items
+
+**Benefit**: Reduces 30 individual items to 5-6 themes, focuses discussion
+
+### Dot Voting Variants
+
+**Standard**: Each person gets 3-5 votes, distribute as they wish (can multi-vote same item)
+
+**Forced distribution**: Must place votes on different items (prevents piling)
+
+**Weighted**: Different color dots for "must discuss" (3 pts) vs "nice to have" (1 pt)
+
+**Quadrant voting**: Vote on impact AND effort separately, plot on 2x2 matrix
+
+### Parking Lot Management
+
+**Purpose**: Capture important but off-topic ideas without derailing
+
+**Process**:
+- Visible "parking lot" section (whiteboard, digital doc)
+- When topic emerges: "This is important. Let's park it and return if time."
+- At end: Review parking lot, decide which to address (next retro, separate meeting, async)
+
+**Don't let parking lot become graveyard**: Follow up or explicitly discard
+
+## 3. Root Cause Analysis Methods
+
+### 5 Whys (Iterative Questioning)
+
+**Process**:
+1. State problem: "We missed sprint goal by 20%"
+2. Why? "Too much unplanned work came in"
+3. Why? "Sales committed features without engineering input"
+4. Why? "No clear process for vetting customer requests"
+5. Why? "We haven't defined roles in customer escalations"
+6. Why? "Product/Sales alignment meetings were cancelled repeatedly"
+
+**Root cause**: Inconsistent cross-functional communication, not just "too much work"
+
+**Pitfalls**: Stopping too early (symptom not root), blaming people not systems
+
+### Fishbone/Ishikawa Diagram (Categorical)
+
+**Structure**: Problem at head, "bones" are categories of causes
+
+**Categories (6Ms for manufacturing, adapt for software)**:
+- **Methods**: Processes, workflows
+- **Machines**: Tools, systems, infrastructure
+- **Materials**: Code, data, resources
+- **Measurements**: Metrics, monitoring
+- **Mother Nature/Environment**: External factors
+- **Manpower/People**: Skills, capacity, communication
+
+**Process**:
+1. Draw fishbone, problem at right
+2. Brainstorm causes in each category
+3. For each cause, ask "why?" to find sub-causes
+4. Identify which causes have most sub-causes or highest impact
+5. Select root causes to address
+
+**When**: Complex problem with multiple contributing factors, want structured brainstorming
+
+### Timeline Analysis (Chronological Reconstruction)
+
+**Process**:
+1. Draw timeline of period (days, weeks, sprints)
+2. Plot events chronologically:
+   - Decisions made
+   - Incidents occurred
+   - Metrics changed
+   - External events (outages, launches, org changes)
+3. Mark sentiment highs/lows
+4. Look for patterns:
+   - What preceded highs? (replicate)
+   - What preceded lows? (avoid)
+   - Clustering of incidents?
+   - Recurring cycles?
+
+**When**: Long period (quarter), complex project, team needs shared understanding of sequence
+
+**Insight example**: "Every time we skip retrospective, next sprint has more bugs" (correlation)
+
+### Systemic Root Cause (Layers of Systems)
+
+**Go beyond immediate cause to systemic**:
+- **Immediate**: "Typo caused outage"
+- **Proximate**: "No code review caught it"
+- **Systemic**: "Code review process not enforced on urgent fixes"
+- **Cultural**: "Urgency culture prioritizes speed over safety"
+
+**Questions to surface systemic issues**:
+- "What allowed this to happen?"
+- "What would prevent this in the future?"
+- "Is this an isolated incident or pattern?"
+- "What incentives/pressures contributed?"
+
+## 4. Psychological Safety Building
+
+### Pre-Retro Norms Setting
+
+**Establish explicitly** (don't assume):
+- Prime Directive: Everyone did best with what they had
+- Focus on systems/processes, not people
+- Listen to understand, not to judge
+- Disagree respectfully
+- What's said here stays here (Chatham House Rule)
+- Assume positive intent
+
+**Post visibly**: On screen share, whiteboard, meeting notes
+
+**Reference during**: "Remember, we're focusing on the process, not individuals"
+
+### Blameless Language
+
+**Shift from**:
+- "You broke production" → "Production broke when X was deployed"
+- "Why didn't you test this?" → "What prevented testing from catching this?"
+- "That was a stupid decision" → "What information led to that decision?"
+
+**Focus on learning, not fault**: "What can we learn?" not "Who's responsible?"
+
+### Retro Facilitator Rotation
+
+**Why**: Prevents one person's biases from dominating, distributes facilitation skill
+
+**How**: Rotate each retro, pair new facilitators with experienced
+
+**Training**: Share facilitation guide, observe retros, debrief after facilitation
+
+### Confidentiality vs Transparency
+
+**Rule of thumb**: What's said in retro stays in retro UNLESS:
+- It's an action item that needs external visibility
+- Team explicitly decides to share broadly
+- It's a safety/legal/ethical issue requiring escalation
+
+**Document actions, not discussions**: Share what we'll do, not who said what
+
+### Building Trust Over Time
+
+**Early retros** (first 3-5): Focus on safe topics (tools, process), celebrate wins, build habit
+
+**Mid-term** (after ~10): Team comfortable, can address harder topics (communication, decision-making)
+
+**Mature retros**: Can discuss interpersonal issues, performance, strategic direction
+
+**Regression**: If trust breaks (leadership change, reorganization), return to basics
+
+## 5. Action Tracking and Metrics
+
+### Action Completion Tracking
+
+**Metrics**:
+- **Completion rate**: % actions completed before next retro (target: >80%)
+- **Cycle time**: Days from action creation to completion (target: <14 days for sprint teams)
+- **Carry-over rate**: % actions moved to next retro (target: <20%)
+
+**Dashboard** (simple spreadsheet or tool):
+
+| Action | Owner | Due | Status | Completed | Notes |
+|--------|-------|-----|--------|-----------|-------|
+| Create standup template | Sarah | Nov 18 | ✓ Done | Nov 17 | |
+| Fix CI pipeline timeout | James | Nov 25 | ⚠ In Progress | - | Blocked on infra |
+
+**Review at start of every retro**: Celebrate completions, discuss blockers, decide on carry-overs
+
+### Leading Indicators (Retro Health)
+
+**Participation rate**: % team attending (target: 100% or explained absence)
+
+**Engagement**: % participants contributing ideas (target: >80%)
+
+**ROTI scores**: % thumbs up on retro value (target: >80%)
+
+**Sentiment trend**: Are Mad/Sad decreasing, Glad increasing over time?
+
+**Repeat issues**: Same problem >3 retros → escalate to leadership (systemic issue beyond team control)
+
+### Lagging Indicators (Business Impact)
+
+**Team performance**:
+- Velocity trend (increasing/stable/decreasing)
+- Bug rate trend (decreasing ideal)
+- On-time delivery % (increasing ideal)
+
+**Team health**:
+- Turnover rate (decreasing ideal)
+- Engagement scores (increasing ideal)
+- Sick leave patterns (stable or decreasing ideal)
+
+**Correlation**: Do teams with regular retros + high action completion have better metrics?
+
+### Action Type Distribution
+
+**Track action categories** over time:
+- **Process**: 40-50% (change how we work)
+- **Technical**: 20-30% (tools, infrastructure, code quality)
+- **Communication**: 15-25% (meetings, documentation, alignment)
+- **Team dynamics**: 5-15% (collaboration, morale, conflict)
+
+**Red flags**:
+- >60% process actions → too much overhead, simplify
+- 0% technical actions → accumulating technical debt
+- >30% communication actions → organizational dysfunction
+
+## 6. Remote and Async Retrospectives
+
+### Synchronous Remote Retros
+
+**Tools**: Miro, Mural, Jamboard, FigJam, Retrium
+
+**Best practices**:
+- **Cameras on**: Increases engagement, reads body language
+- **Silent brainstorming**: Everyone types simultaneously (prevents groupthink, balances introverts/extroverts)
+- **Timers visible**: Keep time-boxing discipline
+- **Breakout rooms**: For large teams, split for intimate discussion, reconvene for synthesis
+- **Emoji reactions**: Quick, non-verbal feedback during shares
+
+**Challenges**:
+- Harder to read room energy → Check in explicitly: "Sensing some tension, am I reading that right?"
+- Tech issues → Have backup plan (phone-in, async fallback)
+- Timezone spread → Rotate times to share burden, or go async
+
+### Async Retrospectives
+
+**When**: Extreme timezone spread (>8 hours), team prefers written reflection
+
+**Process** (5-day cycle):
+1. **Monday**: Facilitator posts retro format, instructions, deadline (Friday)
+2. **Mon-Thu**: Team adds items asynchronously
+3. **Thursday**: Facilitator clusters items, posts summary, asks for votes
+4. **Friday**: Team votes on priorities, facilitator summarizes top themes
+5. **Friday PM**: Facilitator proposes actions based on votes/themes
+6. **Monday (next week)**: Finalize actions, assign owners
+
+**Pros**: Time to reflect deeply, written record, inclusive of all timezones
+**Cons**: Less energy, harder to build on ideas, slower resolution
+
+**Hybrid approach**: Async data gathering, sync discussion and action planning (60 min)
+
+### Tooling Recommendations
+
+**Simple** (small teams, <10 people):
+- Google Docs/Slides: Shared doc, everyone edits simultaneously
+- Jamboard: Simple, sticky-note style
+
+**Advanced** (large teams, dedicated retros):
+- **Retrium**: Purpose-built for retros, many formats, voting, action tracking
+- **Miro/Mural**: Infinite canvas, rich templates, integrations
+- **Parabol**: Open-source, async support, Jira integration
+
+**Action tracking**:
+- Lightweight: Spreadsheet in shared drive
+- Integrated: Jira, Linear, Asana (link retro actions to work tracking)
+- Dedicated: Retrium, Parabol (built-in action tracking and reminders)
+
+---
+
+## Workflow Integration
+
+**When to use advanced techniques**:
+
+**Advanced formats** → When standard formats feel stale, team wants variety
+**Root cause analysis** → When recurring issues appear >3 times, need depth
+**Facilitation techniques** → When participation imbalanced, conflicts emerge
+**Psychological safety building** → New teams, post-conflict, low trust
+**Metrics tracking** → Mature retro practice, want to measure improvement
+**Remote/async** → Distributed teams, timezone challenges
+
+**Progression**:
+1. **Start simple**: Start/Stop/Continue, 30-min retros, basic facilitation
+2. **Build habit**: Consistent schedule, track action completion, >80% attendance
+3. **Deepen practice**: Experiment with formats, root cause techniques, track metrics
+4. **Embed in culture**: Retros feel safe, honest, valuable; team requests retros proactively
--- a/skills/reviews-retros-reflection/resources/template.md
+++ b/skills/reviews-retros-reflection/resources/template.md
@@ -0,0 +1,358 @@
+# Reviews, Retros & Reflection Template
+
+## Workflow
+
+Copy this checklist and track your progress:
+
+```
+Retrospective Progress:
+- [ ] Step 1: Set the stage
+- [ ] Step 2: Gather data
+- [ ] Step 3: Generate insights
+- [ ] Step 4: Decide actions
+- [ ] Step 5: Close retrospective
+```
+
+**Step 1: Set the stage**
+
+Define period, review previous actions, state Prime Directive. See [Setting the Stage](#setting-the-stage).
+
+**Step 2: Gather data**
+
+Collect metrics, events, feedback using chosen format. See [Format Templates](#format-templates).
+
+**Step 3: Generate insights**
+
+Identify patterns, root causes, surprises. See [Insight Generation](#generating-insights).
+
+**Step 4: Decide actions**
+
+Vote and prioritize, create SMART actions. See [Action Planning](#action-planning).
+
+**Step 5: Close**
+
+Commit, schedule follow-up, appreciate. See [Closing](#closing-the-retrospective).
+
+---
+
+## Retrospective Template
+
+### Setting the Stage
+
+**Period**: [Sprint 47 / Q3 2024 / Project Alpha / Week of Nov 11-15]
+
+**Duration**: [Start time - End time, e.g., 2:00-3:30 PM]
+
+**Participants**: [Names and roles]
+
+**Facilitator**: [Name]
+
+**Prime Directive**:
+"Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, and the situation at hand."
+
+**Previous Action Items** (from last retro):
+
+| Action | Owner | Status | Notes |
+|--------|-------|--------|-------|
+| [Previous action 1] | [Name] | ✓ Done / ⚠ In Progress / ✗ Not Done | [Update] |
+| [Previous action 2] | [Name] | [Status] | [Update] |
+
+**Completion rate**: [X/Y completed = Z%]
+
+---
+
+### Format Templates
+
+**Choose one format below based on context:**
+
+#### Option 1: Start/Stop/Continue
+
+**What should we START doing?**
+- [Item 1]: [Why this would help]
+- [Item 2]: [Why this would help]
+
+**What should we STOP doing?**
+- [Item 1]: [Why this isn't working]
+- [Item 2]: [Why this isn't working]
+
+**What should we CONTINUE doing?**
+- [Item 1]: [Why this is working well]
+- [Item 2]: [Why this is working well]
+
+#### Option 2: Mad/Sad/Glad
+
+**What made you MAD?** (frustrated, angry)
+- [Item 1]: [What happened, impact]
+- [Item 2]: [What happened, impact]
+
+**What made you SAD?** (disappointed)
+- [Item 1]: [What happened, impact]
+- [Item 2]: [What happened, impact]
+
+**What made you GLAD?** (happy, satisfied)
+- [Item 1]: [What happened, impact]
+- [Item 2]: [What happened, impact]
+
+#### Option 3: 4Ls
+
+**What did we LOVE?**
+- [Item 1]: [What worked exceptionally well]
+- [Item 2]
+
+**What did we LEARN?**
+- [Item 1]: [New insight or skill gained]
+- [Item 2]
+
+**What did we LACK?**
+- [Item 1]: [Missing resource, skill, or support]
+- [Item 2]
+
+**What did we LONG FOR?**
+- [Item 1]: [What we wish we had]
+- [Item 2]
+
+#### Option 4: Sailboat/Speedboat
+
+**🌬️ WIND** (What's helping us move forward?)
+- [Item 1]: [Accelerator, enabler]
+- [Item 2]
+
+**⚓ ANCHOR** (What's slowing us down?)
+- [Item 1]: [Blocker, drag]
+- [Item 2]
+
+**🪨 ROCKS** (What risks lie ahead?)
+- [Item 1]: [Upcoming risk or danger]
+- [Item 2]
+
+**🏝️ ISLAND** (Where are we heading?)
+- [Goal/vision we're working toward]
+
+#### Option 5: Timeline
+
+**Timeline of events**:
+
+```
+Week 1          Week 2          Week 3          Week 4
+  |               |               |               |
+  📈 High         📉 Low          📈 High         📊 Medium
+
+Key events:
+- [Date]: [Event 1 - why it was a high/low]
+- [Date]: [Event 2]
+- [Date]: [Event 3]
+```
+
+**Patterns observed**:
+- [Pattern 1]: [What we noticed across the timeline]
+- [Pattern 2]
+
+---
+
+### Generating Insights
+
+**Themes and patterns** (group similar items):
+
+**Theme 1: [Name, e.g., "Communication gaps"]**
+- Related items: [Item A, Item B, Item C]
+- Root cause: [Why is this happening? Use 5 Whys if needed]
+- Impact: [How is this affecting team/product/customers?]
+
+**Theme 2: [Name]**
+- Related items: [...]
+- Root cause: [...]
+- Impact: [...]
+
+**Surprises** (unexpected discoveries):
+- [Surprise 1]: [What we didn't expect to learn]
+- [Surprise 2]
+
+**Wins to celebrate** (recognize successes):
+- [Win 1]: [What went well, who contributed]
+- [Win 2]
+
+---
+
+### Action Planning
+
+**Potential actions** (brainstorm):
+
+| Action | Impact (H/M/L) | Effort (H/M/L) | Votes |
+|--------|----------------|----------------|-------|
+| [Action 1] | H | L | 🟢🟢🟢🟢🟢 (5) |
+| [Action 2] | H | M | 🟢🟢🟢 (3) |
+| [Action 3] | M | L | 🟢🟢 (2) |
+| [Action 4] | L | H | 🟢 (1) |
+
+**Selected Actions** (Top 1-3, SMART format):
+
+**Action 1**: [Specific action using active verb]
+- **Measurable**: [How we'll know it's done / success metric]
+- **Assigned**: [Owner name]
+- **Realistic**: [Why this is achievable]
+- **Time-bound**: [Deadline or milestone]
+
+**Action 2**: [Action]
+- **Measurable**: [...]
+- **Assigned**: [...]
+- **Realistic**: [...]
+- **Time-bound**: [...]
+
+**Action 3**: [Action]
+- **Measurable**: [...]
+- **Assigned**: [...]
+- **Realistic**: [...]
+- **Time-bound**: [...]
+
+---
+
+### Closing the Retrospective
+
+**Commitment**:
+- [ ] Team commits to actions above
+- [ ] Owners accept responsibility
+- [ ] Follow-up scheduled: [Date/time]
+
+**Retrospective feedback** (improve the retro itself):
+- **What worked**: [What to keep in next retro]
+- **What to improve**: [How to make next retro better]
+
+**Appreciation** (recognize contributions):
+- **Shoutouts**: [Thank specific people for specific contributions]
+
+**Next retrospective**: [Date/time]
+
+---
+
+## Format Selection Guidance
+
+### When to Use Each Format
+
+**Start/Stop/Continue** → General purpose, balanced feedback, action-oriented
+- ✓ Regular sprint retros
+- ✓ New teams establishing practices
+- ✓ Time-constrained (30 min)
+- ❌ Avoid if need emotional processing or deep reflection
+
+**Mad/Sad/Glad** → Emotion processing, team health, morale
+- ✓ After difficult period (crunch, conflict, failure)
+- ✓ Team experiencing low morale
+- ✓ Surface hidden frustrations
+- ❌ Avoid if already good team dynamics (can create negativity)
+
+**4Ls** → Comprehensive, learning-focused, depth
+- ✓ End of major project/phase
+- ✓ Quarterly/annual reviews
+- ✓ Want thorough reflection
+- ❌ Avoid for regular sprints (too heavy)
+
+**Sailboat** → Visual, strategic, cross-functional
+- ✓ Product/strategy planning
+- ✓ Teams unfamiliar with retros
+- ✓ Distributed teams (visual aids async participation)
+- ❌ Avoid for tactical process issues
+
+**Timeline** → Complex period, shared understanding
+- ✓ Long period (quarter, multi-month project)
+- ✓ Many events, unclear chronology
+- ✓ New team members joining (catch them up)
+- ❌ Avoid for short sprints (overkill)
+
+---
+
+## Action Quality Checklist
+
+**SMART Criteria** (all actions must meet):
+- [ ] **Specific**: Clear verb, what exactly will be done
+- [ ] **Measurable**: Can verify completion objectively
+- [ ] **Assigned**: Named owner (not "team" or "we")
+- [ ] **Realistic**: Achievable with current resources/skills
+- [ ] **Time-bound**: Explicit deadline or milestone
+
+**Good vs Bad Actions**:
+
+❌ **Bad**: "Improve communication"
+- Not specific (what communication?)
+- Not measurable (how measure improvement?)
+- No owner or deadline
+
+✓ **Good**: "Sarah creates #eng-updates Slack channel and posts daily standup summary by 10 AM starting Monday Nov 18"
+- Specific: Create channel, post summary
+- Measurable: Channel exists, daily post by 10 AM
+- Assigned: Sarah
+- Realistic: <15 min/day, uses existing tools
+- Time-bound: Starting Nov 18
+
+**Additional Quality Checks**:
+- [ ] Addresses root cause (not just symptom)
+- [ ] Team has authority to implement (not dependent on external approval)
+- [ ] Concrete first step defined (not just goal)
+- [ ] Success criteria clear (binary done/not done or numeric threshold)
+- [ ] Owner committed and present (not absent team member)
+
+---
+
+## Facilitation Tips
+
+**Setting the Stage** (5-10 min): Read Prime Directive aloud, review previous actions (celebrate completions, decide on incompletes)
+
+**Gathering Data** (15-20 min): Silent writing (5 min) → Share round-robin (10-15 min) → Cluster similar items
+
+**Generating Insights** (15-20 min): Dot voting for focus (5 min) → Discuss top items with 5 Whys (10-15 min)
+
+**Deciding Actions** (10-15 min): Brainstorm solutions (5 min) → Vote and prioritize (5-10 min) → Make SMART, get owner commitment
+
+**Closing** (5 min): Confirm actions, appreciate contributions, retro the retro
+
+---
+
+## Common Patterns by Team Type
+
+**Software Engineering Team**:
+- **Frequency**: Every 2-week sprint
+- **Format**: Start/Stop/Continue
+- **Common themes**: Code quality, technical debt, deployment process, pairing/collaboration
+- **Action examples**: Adopt linter rule, schedule refactoring day, improve CI/CD pipeline
+
+**Product Team**:
+- **Frequency**: After each release or monthly
+- **Format**: 4Ls or Sailboat
+- **Common themes**: User feedback, roadmap clarity, stakeholder communication, scope creep
+- **Action examples**: User interview cadence, stakeholder update format, prioritization framework
+
+**Marketing/Growth Team**:
+- **Frequency**: Weekly or bi-weekly
+- **Format**: Wins/Blockers/Priorities (custom)
+- **Common themes**: Campaign performance, cross-functional dependencies, experiment velocity
+- **Action examples**: A/B test process, creative review workflow, metrics dashboard
+
+**Leadership/Executive Team**:
+- **Frequency**: Quarterly
+- **Format**: Timeline + 4Ls
+- **Common themes**: Strategic alignment, resource allocation, organizational health, market changes
+- **Action examples**: OKR calibration, communication forums, talent strategy
+
+---
+
+## Quality Checklist
+
+### Before the Retro
+- [ ] Scheduled with sufficient notice (≥3 days)
+- [ ] Participants know format and expectations
+- [ ] Previous action items reviewed
+- [ ] Metrics/data gathered if relevant
+
+### During the Retro
+- [ ] Prime Directive stated upfront
+- [ ] All voices heard (facilitation balanced participation)
+- [ ] Time-boxed (started and ended on schedule)
+- [ ] Discussion stayed blameless and constructive
+- [ ] Actions are SMART (specific, measurable, assigned, realistic, time-bound)
+- [ ] 1-3 actions maximum (not overwhelming)
+
+### After the Retro
+- [ ] Actions documented and shared with team
+- [ ] Owners confirmed and committed
+- [ ] Follow-up scheduled (date in calendar)
+- [ ] Next retro scheduled
+- [ ] Completion tracked (>80% target before next retro)