Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Ask happy clients for a testimonial or review the right way — timed well, easy to give, and specific enough to actually persuade future clients. Use when asked to get testimonials, ask for a review, how to request a testimonial, or get social proof from clients. Produces the right moment and channel to ask, a message that makes saying yes effortless, guiding questions that yield specific results-based testimonials (not 'great to work with'), how to handle a written vs video ask, and permission/u
.claude/skills/mohitagw15856-testimonial-request/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 75% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 62% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 11% | 0% |
Testimonials are the highest-converting marketing you have, and most freelancers/businesses never collect them — or collect vague "they were great!" ones that persuade no one. The fix is asking at the right moment, making it effortless, and guiding the client toward a specific, results-focused answer. This does that without feeling like you're fishing for compliments.
Ask for these if not provided:
When/where to ask: peak-goodwill moment · channel]. The message > Warm, specific to the result, effortless ask + guiding questions or an offer to draft, and where it'll help].
Guiding questions (for specifics): "what was the problem before?" · "what changed / what result?" · "what would you tell someone considering me?"]. Format note: written vs video — the ask + effort]. Permission: OK to use with name/role/company on placement]]. Tone: appreciative, not needy; one gentle follow-up max.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,201 | 22,467 | +31% | 1 | 1 | 0% | 2,661 | 3,454 | +30% | 0 | 0 | — |
case-02 | fail→fail | 15,518 | 19,622 | +26% | 1 | 1 | 0% | 2,466 | 3,346 | +36% | 0 | 0 | — |
case-03 | fail→pass | 20,622 | 17,967 | -13% | 1 | 1 | 0% | 3,183 | 3,457 | +9% | 0 | 0 | — |
case-04 | fail→pass | 13,726 | 16,473 | +20% | 1 | 1 | 0% | 1,438 | 2,522 | +75% | 0 | 0 | — |
case-05 | fail→pass | 16,987 | 16,311 | -4% | 1 | 1 | 0% | 1,583 | 2,569 | +62% | 0 | 0 | — |
case-06 | fail→pass | 25,480 | 17,356 | -32% | 1 | 1 | 0% | 2,657 | 2,783 | +5% | 0 | 0 | — |
case-07 | fail→pass | 14,860 | 15,666 | +5% | 1 | 1 | 0% | 2,265 | 2,517 | +11% | 0 | 0 | — |
case-08 | fail→pass | 16,217 | 17,076 | +5% | 1 | 1 | 0% | 2,081 | 2,608 | +25% | 0 | 0 | — |
case-09 | fail→fail | 15,787 | 15,472 | -2% | 1 | 1 | 0% | 1,487 | 2,484 | +67% | 0 | 0 | — |
case-10 | fail→fail | 27,602 | 12,561 | -54% | 1 | 1 | 0% | 2,210 | 2,690 | +22% | 0 | 0 | — |
case-11 | fail→pass | 16,415 | 15,033 | -8% | 1 | 1 | 0% | 1,490 | 2,563 | +72% | 0 | 0 | — |
case-12 | fail→fail | 14,950 | 16,289 | +9% | 1 | 1 | 0% | 1,493 | 2,474 | +66% | 0 | 0 | — |
case-13 | fail→pass | 12,444 | 17,236 | +39% | 1 | 1 | 0% | 1,534 | 2,517 | +64% | 0 | 0 | — |
case-14 | pass→pass | 17,527 | 16,246 | -7% | 1 | 1 | 0% | 1,543 | 2,619 | +70% | 0 | 0 | — |
case-15 | fail→pass | 13,416 | 11,452 | -15% | 1 | 1 | 0% | 1,500 | 2,696 | +80% | 0 | 0 | — |
case-16 | pass→pass | 10,715 | 16,323 | +52% | 1 | 1 | 0% | 1,512 | 2,317 | +53% | 0 | 0 | — |
case-17 | fail→fail | 16,211 | 13,404 | -17% | 1 | 1 | 0% | 1,480 | 2,263 | +53% | 0 | 0 | — |
case-18 | fail→pass | 17,873 | 12,219 | -32% | 1 | 1 | 0% | 1,635 | 2,312 | +41% | 0 | 0 | — |
case-19 | pass→fail | 67,881 | 24,803 | -63% | 1 | 1 | 0% | 8,233 | 4,093 | -50% | 0 | 0 | — |
case-20 | pass→pass | 23,381 | 31,863 | +36% | 1 | 1 | 0% | 2,520 | 4,246 | +68% | 0 | 0 | — |
case-21 | pass→pass | 22,401 | 10,887 | -51% | 1 | 1 | 0% | 1,642 | 2,341 | +43% | 0 | 0 | — |
case-22 | pass→fail | 22,191 | 27,653 | +25% | 1 | 1 | 0% | 2,210 | 4,294 | +94% | 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. The headline lift of +36 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are 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.