Session Insights

# Session history and learnings
cat .claude/LEARNED.md 2>/dev/null || cat CLAUDE.md | grep -A999 "LEARNED"
cat .claude/learning-log.md 2>/dev/null

# Session activity
git log --oneline --since="today" --author="$(git config user.name)"
git diff --stat

Session Insights
  Duration: 47 min
  Edits: 23 files modified
  Corrections: 2 self-corrections applied
  Learnings: 3 new patterns captured
  Context: 62% used (safe)

Learning Insights (42 total)

Top categories:
  Testing     12 learnings (29%)
  Navigation   8 learnings (19%)
  Git          7 learnings (17%)
  Quality      6 learnings (14%)

Most applied:
  #12 [Testing] Run tests before commit — 15 times
  #8  [Navigation] Confirm path for common names — 11 times
  #3  [Git] Always pull before creating branches — 9 times

Stale learnings (never applied):
  #15 [Editing] Prefer named exports — 0 times (45 days old)
  #22 [Quality] Use strict TypeScript config — 0 times (30 days old)
  
  → Consider: Review if these patterns are still relevant

Correction Heatmap

By category (all time):
  ████████████ Testing      34 corrections
  ████████     Navigation   22 corrections
  ██████       Git          18 corrections
  ████         Quality      12 corrections
  ██           Architecture  6 corrections
  █            Performance   3 corrections

Hot learnings (most corrected, least learned):
  - [Testing] Mock external deps — corrected 8x, learned 0x
    → Consider: /learn-rule to capture this permanently
    
  - [Navigation] Check src/lib/ before src/utils/ — corrected 5x, learned 0x
    → Consider: Add to LEARNED section

Cold learnings (learned but never applied):
  - [Editing] Use named exports — learned 45 days ago, applied 0x
    → Consider: Remove if no longer relevant
    
  - [Architecture] Avoid default exports — learned 30 days ago, applied 0x
    → Consider: Review if pattern changed

Productivity (last 10 sessions)
  Avg session: 35 min
  Avg edits/session: 18 files
  Correction rate: 12% (improving ↓)
  Learning capture: 2.1 per session
  
Trend:
  Week 1: 18% correction rate
  Week 2: 15% correction rate
  Week 3: 12% correction rate
  
  ✓ Correction rate decreasing — learnings are working!
  
Top productivity days:
  - 2026-03-05: 8 edits, 0 corrections (0% rate) ✓
  - 2026-03-01: 12 edits, 1 correction (8.3% rate) ✓
  - 2026-02-28: 15 edits, 1 correction (6.7% rate) ✓

=====================================
SESSION INSIGHTS
=====================================

CURRENT SESSION
  Duration: 47 min
  Files edited: 23
  Corrections: 2 (8.7% rate)
  Learnings captured: 3
  Context usage: 62%

LEARNING ANALYTICS (42 total learnings)

Top categories:
  █████████ Testing     12 (29%)
  ██████       Navigation   8 (19%)
  █████         Git          7 (17%)
  ████          Quality      6 (14%)

Most applied learnings:
  1. [Testing] Run tests before commit — 15x
  2. [Navigation] Confirm path for common names — 11x
  3. [Git] Always pull before creating branches — 9x
  4. [Testing] Mock external APIs — 8x
  5. [Quality] Remove console.log before commit — 7x

Stale learnings (never applied):
  - [Editing] Prefer named exports (45 days old)
  - [Quality] Use strict TypeScript config (30 days old)
  
  → Review and update or remove these patterns

CORRECTION HEATMAP

By category (all time):
  ████████████ Testing      34 corrections
  ████████     Navigation   22 corrections
  ██████       Git          18 corrections
  ████         Quality      12 corrections

Hot learnings (corrected often, not learned):
  - [Testing] Mock external deps — 8 corrections, 0 learnings
  - [Navigation] Check src/lib/ first — 5 corrections, 0 learnings
  
  → Use /learn-rule to capture these patterns

PRODUCTIVITY TRENDS (last 10 sessions)

Average session: 35 min, 18 edits
Correction rate: 12% (improving ↓)
  Week 1: 18%
  Week 2: 15%
  Week 3: 12%
  
  ✓ Learnings are reducing mistakes!

Best sessions (lowest correction rate):
  1. 2026-03-05: 8 edits, 0 corrections (0%)
  2. 2026-03-01: 12 edits, 1 correction (8.3%)
  3. 2026-02-28: 15 edits, 1 correction (6.7%)

Worst sessions (highest correction rate):
  1. 2026-02-20: 10 edits, 4 corrections (40%)
     Topic: React performance optimization
     → High correction area — add more learnings
     
  2. 2026-02-15: 8 edits, 3 corrections (37.5%)
     Topic: Database migrations
     → Review patterns before next migration

=====================================
SUGGESTIONS

1. Capture "Mock external deps" as learning (corrected 8x, never learned)
2. Review stale learnings: named exports, strict TS (never applied)
3. React performance is high correction area (40% rate) — add more patterns
4. ✓ Overall trend improving — keep capturing learnings!

=====================================

## Session Tracking

After each session:
- Track correction count
- Log session duration and edits
- Note topics and correction categories
- Store in .claude/sessions/[date].json

~/.pro-workflow/data.db
  ├── learnings (id, category, rule, created, applied_count)
  ├── sessions (id, date, duration, edits, corrections, topics)
  └── corrections (id, session_id, category, learning_id)

/insights --last 30
/insights --category Testing
/insights --topic "database migrations"

SELECT topic, 
       COUNT(*) as sessions,
       AVG(corrections * 100.0 / edits) as avg_correction_rate
FROM sessions
GROUP BY topic
ORDER BY avg_correction_rate DESC;

SELECT l.category, l.rule, l.applied_count,
       (SELECT COUNT(*) FROM corrections c 
        WHERE c.category = l.category 
        AND c.created < l.created) as corrections_before
FROM learnings l
WHERE l.applied_count > 5
ORDER BY l.applied_count DESC;

SELECT DATE(created) as date,
       AVG(corrections * 100.0 / edits) as correction_rate
FROM sessions
WHERE created > DATE('now', '-30 days')
GROUP BY date
ORDER BY date;