Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Capture a life story — your own or a parent's/grandparent's — into a keepsake, using good interview questions and a structure that turns memories into readable stories. Use when asked to help write a memoir, capture my parent's/grandparent's story, record family history, or preserve someone's life story. Produces a question set that unlocks real memories (not just dates), a session plan for interviewing over time, a structure to organize stories into chapters or themes, prompts to draw out detai
.claude/skills/mohitagw15856-memoir-story-capture/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 73% | 0% |
Every family loses irreplaceable stories when someone passes — the memories that were never written down. This helps you capture them: the questions that unlock real, vivid memories (not a dry timeline), a way to interview over manageable sessions, and a structure to turn the raw stories into something readable and keepable — a keepsake for the family before the chance is gone.
Ask for these if not provided:
Question set (unlocks memories): childhood/sensory · turning points · people/relationships · hardships · joys · everyday life · hopes & regrets · advice]. Session plan: how many, how long, recorded how] — capture raw first if time-sensitive. Draw-out follow-ups: "what did that feel like?" · "who was there?" · "what happened next?"]. Structure: chapters/themes or chronology]. Final form: memoir / transcript / recording / photo-story book]. Preserve: record + back up the raw material].
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 42,066 | 32,310 | -23% | 1 | 1 | 0% | 5,897 | 5,077 | -14% | 0 | 0 | — |
case-02 | fail→pass | 40,597 | 26,366 | -35% | 1 | 1 | 0% | 5,481 | 3,607 | -34% | 0 | 0 | — |
case-03 | fail→pass | 32,808 | 20,653 | -37% | 1 | 1 | 0% | 4,228 | 4,287 | +1% | 0 | 0 | — |
case-04 | pass→pass | 15,173 | 19,906 | +31% | 1 | 1 | 0% | 2,427 | 3,418 | +41% | 0 | 0 | — |
case-05 | fail→pass | 24,937 | 20,597 | -17% | 1 | 1 | 0% | 3,297 | 3,675 | +11% | 0 | 0 | — |
case-06 | fail→pass | 23,954 | 20,426 | -15% | 1 | 1 | 0% | 2,879 | 3,538 | +23% | 0 | 0 | — |
case-07 | fail→pass | 17,881 | 23,427 | +31% | 1 | 1 | 0% | 2,118 | 3,665 | +73% | 0 | 0 | — |
case-08 | fail→pass | 14,949 | 20,329 | +36% | 1 | 1 | 0% | 2,403 | 3,482 | +45% | 0 | 0 | — |
case-09 | fail→fail | 15,564 | 23,961 | +54% | 1 | 1 | 0% | 2,511 | 3,848 | +53% | 0 | 0 | — |
case-10 | pass→pass | 19,672 | 22,386 | +14% | 1 | 1 | 0% | 2,484 | 3,680 | +48% | 0 | 0 | — |
case-11 | fail→fail | 16,424 | 14,710 | -10% | 1 | 1 | 0% | 1,879 | 3,362 | +79% | 0 | 0 | — |
case-12 | fail→fail | 22,342 | 23,214 | +4% | 1 | 1 | 0% | 2,640 | 3,794 | +44% | 0 | 0 | — |
case-13 | fail→fail | 18,012 | 13,903 | -23% | 1 | 1 | 0% | 2,198 | 3,468 | +58% | 0 | 0 | — |
case-14 | pass→pass | 14,232 | 22,857 | +61% | 1 | 1 | 0% | 2,385 | 3,968 | +66% | 0 | 0 | — |
case-15 | pass→pass | 24,222 | 20,345 | -16% | 1 | 1 | 0% | 2,535 | 3,601 | +42% | 0 | 0 | — |
case-16 | pass→pass | 20,938 | 18,988 | -9% | 1 | 1 | 0% | 2,371 | 4,103 | +73% | 0 | 0 | — |
case-17 | pass→pass | 11,483 | 15,504 | +35% | 1 | 1 | 0% | 1,841 | 2,598 | +41% | 0 | 0 | — |
case-18 | pass→pass | 20,507 | 22,179 | +8% | 1 | 1 | 0% | 2,488 | 4,211 | +69% | 0 | 0 | — |
case-19 | pass→pass | 16,782 | 17,348 | +3% | 1 | 1 | 0% | 1,745 | 2,835 | +62% | 0 | 0 | — |
case-20 | pass→pass | 17,672 | 22,791 | +29% | 1 | 1 | 0% | 2,155 | 4,165 | +93% | 0 | 0 | — |
case-21 | pass→pass | 27,055 | 28,294 | +5% | 1 | 1 | 0% | 3,573 | 4,563 | +28% | 0 | 0 | — |
case-22 | pass→pass | 26,471 | 24,975 | -6% | 1 | 1 | 0% | 3,271 | 3,814 | +17% | 0 | 0 | — |
case-23 | pass→pass | 10,933 | 11,497 | +5% | 1 | 1 | 0% | 1,178 | 2,135 | +81% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases.
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.