home / skills / alienfast / claude / retro
This skill analyzes completed sprints to reveal velocity trends, scope changes, and patterns for actionable retro improvements.
npx playbooks add skill alienfast/claude --skill retroReview the files below or copy the command above to add this skill to your agents.
---
name: retro
description: Analyze completed Linear cycles for retrospectives. Identifies velocity trends, scope creep, and patterns to improve future planning.
---
# Retro Skill - Sprint Retrospective
You are an expert at analyzing completed sprints and generating actionable retrospective insights.
## When to Use
Use this skill when:
- A sprint/cycle has ended
- Preparing for a retrospective meeting
- Analyzing team performance trends
## Process
1. **Gather Cycle Data**
```bash
linear cycles analyze --team ENG --count 3
linear cycles get CYCLE-ID
```
2. **Analyze Completion**
- Planned vs actual completion
- Scope changes during cycle
- Carry-over items
3. **Identify Patterns**
- Estimation accuracy
- Blocker frequency
- Category distribution
4. **Generate Insights**
## Analysis Areas
### Velocity & Completion
- Points planned vs completed
- Issue count planned vs completed
- Completion rate trend
### Scope Management
- Issues added mid-cycle
- Issues removed/deferred
- Scope creep percentage
### Blockers & Dependencies
- Issues that were blocked
- Average time blocked
- Dependency-related delays
### Workload Distribution
- Points per team member
- Completion rate by assignee
- Bottleneck identification
## Output Format
```
RETROSPECTIVE: Cycle 23
════════════════════════════════════════
SUMMARY
────────────────────────────────────────
Planned: 40 points (12 issues)
Completed: 34 points (10 issues)
Completion Rate: 85%
SCOPE CHANGES
────────────────────────────────────────
Added mid-cycle: 3 issues (8 pts)
Removed: 1 issue (2 pts)
Carried over: 2 issues (6 pts)
Scope creep: 15%
WHAT WENT WELL
────────────────────────────────────────
+ Auth refactor completed ahead of schedule
+ Zero critical bugs in production
+ Good collaboration on API work
WHAT COULD IMPROVE
────────────────────────────────────────
- 2 issues blocked for 3+ days (ENG-201, ENG-203)
- Underestimated ENG-205 (3pts planned, 8pts actual)
- Late scope additions disrupted focus
TRENDS (3 cycles)
────────────────────────────────────────
Velocity: 34 → 32 → 34 (stable)
Completion: 85% → 88% → 85% (stable)
Scope creep: 12% → 8% → 15% (increasing)
ACTION ITEMS
────────────────────────────────────────
1. Add buffer for complex API work
2. Daily standup check on blockers
3. Scope freeze after day 2
```
## Commands Used
```bash
# Analyze recent cycles
linear cycles analyze --team ENG --count 3
# Get specific cycle details
linear cycles get CYCLE-23
# List cycle issues
linear issues list --team ENG --cycle CYCLE-23
# Check what was blocked
linear deps --team ENG
```
## Discussion Questions
Generate discussion prompts for the team:
1. **Wins**: What should we celebrate?
2. **Blockers**: What slowed us down?
3. **Process**: What would we do differently?
4. **Tools**: Are there tools/automation to add?
5. **Communication**: Where did we lose sync?
## Best Practices
1. **Be data-driven** - Base insights on actual metrics
2. **Focus on process** - Not individual performance
3. **Actionable items** - Every insight should have a follow-up
4. **Track over time** - Compare to previous retros
This skill analyzes completed Linear cycles to generate actionable retrospective insights. It identifies velocity trends, scope creep, blockers, and workload distribution to improve future sprint planning. The output is concise, data-driven, and structured for quick discussion and follow-up.
The skill gathers cycle metadata and issue lists, compares planned versus completed points and issue counts, and detects mid-cycle scope changes. It highlights blocked work, estimation gaps, and per-assignee load, then summarizes trends across recent cycles and produces prioritized action items and discussion prompts.
What inputs does the skill need?
Cycle identifiers or a count of recent cycles and team context to fetch issue lists, estimates, and timeline events.
How are scope creep and blockers measured?
Scope creep is percentage of points or issues added mid-cycle; blockers are derived from dependency/blocked flags and time-in-blocked-state metrics.