Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get the single honest thing you're avoiding about a situation — said kindly but not softened away. Use when asked tell me the hard truth, what am I avoiding here, be honest with me about this, or what do I not want to hear. Produces the one thing you already half-know but keep sidestepping, said plainly and with care (not cruelty), why it's hard to face, and what facing it would actually make possible — because the truth you're avoiding is usually the one that would change things, and a kind voi
.claude/skills/mohitagw15856-one-hard-truth/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -17% | 0% |
There's usually one thing you already sense but keep steering around — about the relationship, the job, the habit, the plan. Everyone around you is too polite to say it, and you're too close to face it. This says the one hard truth plainly and kindly: not to hurt you, but because naming it is what unlocks the situation. Kind and honest aren't opposites — cruelty and cowardice are the opposites; this is the third thing.
Ask for these if not provided:
The one hard truth: the single honest thing, plainly]. Why it's hard to face: the real cost — change, a story dented, something given up]. But here's what facing it opens: what becomes possible]. One first move: a small step facing it makes available].
(Said because it might help — not to be harsh.)
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 18,087 | 14,736 | -19% | 1 | 1 | 0% | 1,986 | 2,414 | +22% | 0 | 0 | — |
case-02 | pass→pass | 19,670 | 15,763 | -20% | 1 | 1 | 0% | 2,296 | 2,247 | -2% | 0 | 0 | — |
case-03 | pass→pass | 16,185 | 14,924 | -8% | 1 | 1 | 0% | 1,761 | 2,376 | +35% | 0 | 0 | — |
case-04 | fail→pass | 23,174 | 13,034 | -44% | 1 | 1 | 0% | 2,325 | 2,211 | -5% | 0 | 0 | — |
case-05 | fail→pass | 14,109 | 15,227 | +8% | 1 | 1 | 0% | 2,292 | 2,259 | -1% | 0 | 0 | — |
case-06 | fail→pass | 18,138 | 12,014 | -34% | 1 | 1 | 0% | 1,788 | 2,065 | +15% | 0 | 0 | — |
case-07 | fail→pass | 26,166 | 18,546 | -29% | 1 | 1 | 0% | 3,448 | 2,529 | -27% | 0 | 0 | — |
case-08 | fail→pass | 22,204 | 13,511 | -39% | 1 | 1 | 0% | 2,844 | 2,358 | -17% | 0 | 0 | — |
case-09 | fail→pass | 20,712 | 8,809 | -57% | 1 | 1 | 0% | 2,371 | 2,286 | -4% | 0 | 0 | — |
case-10 | fail→pass | 12,243 | 16,715 | +37% | 1 | 1 | 0% | 1,990 | 2,392 | +20% | 0 | 0 | — |
case-11 | fail→pass | 19,094 | 15,236 | -20% | 1 | 1 | 0% | 1,887 | 2,512 | +33% | 0 | 0 | — |
case-12 | pass→pass | 22,930 | 14,605 | -36% | 1 | 1 | 0% | 2,559 | 2,178 | -15% | 0 | 0 | — |
case-13 | fail→pass | 23,160 | 14,884 | -36% | 1 | 1 | 0% | 3,120 | 2,300 | -26% | 0 | 0 | — |
case-14 | pass→fail | 23,712 | 22,924 | -3% | 1 | 1 | 0% | 3,034 | 3,231 | +6% | 0 | 0 | — |
case-15 | pass→pass | 29,480 | 17,522 | -41% | 1 | 1 | 0% | 3,442 | 3,117 | -9% | 0 | 0 | — |
case-16 | pass→fail | 26,313 | 14,064 | -47% | 1 | 1 | 0% | 3,303 | 2,984 | -10% | 0 | 0 | — |
case-17 | fail→pass | 25,800 | 11,399 | -56% | 1 | 1 | 0% | 2,685 | 2,415 | -10% | 0 | 0 | — |
case-18 | fail→pass | 25,450 | 16,543 | -35% | 1 | 1 | 0% | 2,711 | 2,972 | +10% | 0 | 0 | — |
case-19 | fail→pass | 25,367 | 16,260 | -36% | 1 | 1 | 0% | 3,115 | 2,357 | -24% | 0 | 0 | — |
case-20 | fail→pass | 25,811 | 14,293 | -45% | 1 | 1 | 0% | 3,675 | 2,351 | -36% | 0 | 0 | — |
case-21 | fail→pass | 30,973 | 9,683 | -69% | 1 | 1 | 0% | 4,093 | 2,388 | -42% | 0 | 0 | — |
case-22 | fail→pass | 17,381 | 15,472 | -11% | 1 | 1 | 0% | 1,725 | 2,348 | +36% | 0 | 0 | — |
case-23 | pass→pass | 18,074 | 16,047 | -11% | 1 | 1 | 0% | 2,666 | 2,362 | -11% | 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 +57 percentage points is the difference between those two pass rates over the 23 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.