Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze sprint retrospectives for patterns and action item tracking. Usage: /retro analyze <retro_data.json>
.claude/skills/alirezarezvani-retro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 127% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -66% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 445% | 0% |
Analyze retrospective data for recurring themes, sentiment trends, and action item effectiveness.
/retro analyze <retro_data.json> Full retrospective analysisjson{ "sprint_name": "Sprint 24", "went_well": ["CI pipeline improvements", "Pair programming sessions"], "improvements": ["Too many meetings", "Flaky integration tests"], "action_items": [ {"description": "Reduce standup to 10 min", "owner": "SM", "status": "done"}, {"description": "Fix flaky tests", "owner": "QA Lead", "status": "in_progress"} ], "participants": 8 }
/retro analyze sprint-24-retro.json
/retro analyze sprint-24-retro.json --format jsonproject-management/skills/scrum-master/scripts/retrospective_analyzer.py — Retrospective analyzer (<data_file> [--format text|json])> project-management/skills/scrum-master/SKILL.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 3,974 | 1,225 | -69% | 1 | 1 | 0% | 702 | 462 | -34% | 0 | 0 | — |
case-01 | fail→pass | 3,282 | 5,489 | +67% | 1 | 1 | 0% | 550 | 1,251 | +127% | 0 | 0 | — |
case-02 | fail→fail | 4,471 | 12,744 | +185% | 1 | 1 | 0% | 815 | 1,757 | +116% | 0 | 0 | — |
case-03 | pass→pass | 6,855 | 3,802 | -45% | 1 | 1 | 0% | 1,016 | 1,057 | +4% | 0 | 0 | — |
case-04 | pass→fail | 9,169 | 5,251 | -43% | 1 | 1 | 0% | 1,467 | 638 | -57% | 0 | 0 | — |
case-06 | fail→pass | 8,584 | 1,462 | -83% | 1 | 1 | 0% | 1,638 | 555 | -66% | 0 | 0 | — |
case-07 | fail→fail | 8,568 | 4,554 | -47% | 1 | 1 | 0% | 1,463 | 1,131 | -23% | 0 | 0 | — |
case-08 | fail→fail | 7,753 | 10,484 | +35% | 1 | 1 | 0% | 1,295 | 2,028 | +57% | 0 | 0 | — |
case-09 | fail→fail | 10,939 | 4,768 | -56% | 1 | 1 | 0% | 1,869 | 534 | -71% | 0 | 0 | — |
case-10 | fail→fail | 7,023 | 6,705 | -5% | 1 | 1 | 0% | 1,160 | 1,412 | +22% | 0 | 0 | — |
case-11 | fail→fail | 11,870 | 12,578 | +6% | 1 | 1 | 0% | 2,082 | 2,593 | +25% | 0 | 0 | — |
case-12 | fail→pass | 7,064 | 6,613 | -6% | 1 | 1 | 0% | 1,225 | 874 | -29% | 0 | 0 | — |
case-13 | pass→pass | 7,604 | 2,088 | -73% | 1 | 1 | 0% | 1,319 | 589 | -55% | 0 | 0 | — |
case-14 | fail→pass | 4,913 | 7,341 | +49% | 1 | 1 | 0% | 275 | 1,499 | +445% | 0 | 0 | — |
case-15 | fail→fail | 10,266 | 6,891 | -33% | 1 | 1 | 0% | 1,506 | 772 | -49% | 0 | 0 | — |
case-16 | fail→pass | 2,964 | 1,578 | -47% | 1 | 1 | 0% | 420 | 458 | +9% | 0 | 0 | — |
case-17 | pass→pass | 6,153 | 2,153 | -65% | 1 | 1 | 0% | 1,090 | 687 | -37% | 0 | 0 | — |
case-18 | fail→fail | 3,156 | 4,227 | +34% | 1 | 1 | 0% | 521 | 545 | +5% | 0 | 0 | — |
case-19 | fail→fail | 4,890 | 11,494 | +135% | 1 | 1 | 0% | 242 | 1,914 | +691% | 0 | 0 | — |
case-20 | pass→pass | 10,045 | 10,235 | +2% | 1 | 1 | 0% | 1,605 | 1,870 | +17% | 0 | 0 | — |
case-21 | pass→pass | 9,893 | 9,551 | -3% | 1 | 1 | 0% | 1,866 | 2,073 | +11% | 0 | 0 | — |
case-22 | pass→pass | 7,568 | 4,942 | -35% | 1 | 1 | 0% | 1,319 | 1,233 | -7% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 16 counted toward the lift figure. The other 6 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 16 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.