MansoorMajeed/retro

Retrospective

Analyze recent work patterns to find what's working and what isn't. Default: last 14 days.

Data Collection

Run these to gather data:

# Commits with stats
git log --since="14 days ago" --oneline --stat --no-merges

# Commit count per day
git log --since="14 days ago" --format="%ad" --date=short --no-merges | sort | uniq -c

# Most changed files (hotspots)
git log --since="14 days ago" --name-only --no-merges --format="" | sort | uniq -c | sort -rn | head -20

# Commit type breakdown
git log --since="14 days ago" --oneline --no-merges

Analysis

Commit Patterns

• Total commits, LOC added/removed
• Commits per day — any gaps? Bursts?
• Commit type breakdown: feat / fix / refactor / test / chore / docs
• Fix ratio: If fixes are more than 40% of commits, we're shipping too many bugs

Hotspots

• Which files changed most? Frequent changes = instability or active development
• Are test files keeping pace with source files?

Plans vs Reality

• Read recent plans from docs/plans/
• Did they survive contact with reality? What changed?
• Were estimates (number of tasks, scope) accurate?

Tests

• Test LOC vs production LOC ratio
• Are tests being written alongside code or added later?

Output

Write the retrospective to docs/retros/YYYY-MM-DD-retro.md:

# Retrospective: YYYY-MM-DD

**Period:** [start date] to [end date]
**Commits:** N | **LOC:** +X / -Y | **Fix ratio:** Z%

## What Went Well
- [specific observations with evidence]

## What Didn't Go Well
- [specific observations with evidence]

## Hotspots
| File | Changes | Notes |
|------|---------|-------|
| ... | N | ... |

## Plans vs Reality
- [plan name]: [on track / deviated because...]

## Action Items
- [ ] [specific, actionable improvements for next period]

Trend Comparison

If prior retros exist in docs/retros/, compare:

• Is the fix ratio improving or worsening?
• Are hotspots stabilizing or growing?
• Are action items from previous retros being addressed?

Be candid and specific. Use evidence from the git history, not vague impressions.