Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Show interactive dialogs for user input — text, confirmations, choices, date/time pickers. Use when you need user input beyond chat.
.claude/skills/mikeyobrien-dialog/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -41% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -46% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -22% | 0% |
bashtermux-dialog text -t "Title" -i "Hint text" termux-dialog text -m # multiline termux-dialog text -p # password (hidden) termux-dialog text -n # numbers only
bashtermux-dialog confirm -t "Are you sure?"
bashtermux-dialog radio -t "Choose one" -v "Option A,Option B,Option C"
bashtermux-dialog checkbox -t "Select all that apply" -v "Item 1,Item 2,Item 3"
bashtermux-dialog spinner -t "Select" -v "Apple,Banana,Cherry"
bashtermux-dialog sheet -t "Pick" -v "Action 1,Action 2,Action 3"
bashtermux-dialog counter -t "Pick a number" -r "0,100,50" # min,max,start
bashtermux-dialog date -t "Select date" termux-dialog date -d "yyyy-MM-dd" # custom format
bashtermux-dialog time -t "Select time"
bashtermux-dialog speech -t "Speak now"
json{ "code": 0, "text": "user input here" }
code: 0 = OK, code: -1 = canceled"values" array instead of "text"| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 7,481 | 9,050 | +21% | 1 | 1 | 0% | 1,364 | 800 | -41% | 0 | 0 | — |
case-02 | pass→pass | 7,606 | 2,245 | -70% | 1 | 1 | 0% | 1,469 | 786 | -46% | 0 | 0 | — |
case-03 | pass→pass | 4,960 | 1,837 | -63% | 1 | 1 | 0% | 923 | 720 | -22% | 0 | 0 | — |
case-04 | pass→pass | 9,176 | 4,094 | -55% | 1 | 1 | 0% | 1,575 | 1,162 | -26% | 0 | 0 | — |
case-05 | pass→pass | 5,254 | 2,266 | -57% | 1 | 1 | 0% | 878 | 817 | -7% | 0 | 0 | — |
case-06 | pass→pass | 6,297 | 1,719 | -73% | 1 | 1 | 0% | 1,201 | 675 | -44% | 0 | 0 | — |
case-07 | pass→pass | 6,523 | 3,266 | -50% | 1 | 1 | 0% | 1,250 | 1,010 | -19% | 0 | 0 | — |
case-08 | pass→pass | 6,164 | 1,977 | -68% | 1 | 1 | 0% | 1,171 | 815 | -30% | 0 | 0 | — |
case-09 | pass→pass | 6,742 | 1,969 | -71% | 1 | 1 | 0% | 1,299 | 779 | -40% | 0 | 0 | — |
case-10 | pass→pass | 5,470 | 2,739 | -50% | 1 | 1 | 0% | 1,047 | 865 | -17% | 0 | 0 | — |
case-11 | pass→pass | 8,276 | 2,947 | -64% | 1 | 1 | 0% | 1,512 | 942 | -38% | 0 | 0 | — |
case-12 | pass→pass | 9,912 | 2,341 | -76% | 1 | 1 | 0% | 1,845 | 827 | -55% | 0 | 0 | — |
case-13 | pass→pass | 5,316 | 1,711 | -68% | 1 | 1 | 0% | 959 | 668 | -30% | 0 | 0 | — |
case-14 | fail→pass | 6,470 | 2,393 | -63% | 1 | 1 | 0% | 1,184 | 833 | -30% | 0 | 0 | — |
case-15 | fail→pass | 3,471 | 1,678 | -52% | 1 | 1 | 0% | 647 | 652 | +1% | 0 | 0 | — |
case-16 | pass→pass | 5,080 | 1,242 | -76% | 1 | 1 | 0% | 864 | 602 | -30% | 0 | 0 | — |
case-17 | pass→pass | 5,128 | 2,142 | -58% | 1 | 1 | 0% | 970 | 707 | -27% | 0 | 0 | — |
case-18 | pass→pass | 5,175 | 3,127 | -40% | 1 | 1 | 0% | 900 | 775 | -14% | 0 | 0 | — |
case-19 | pass→pass | 4,201 | 1,904 | -55% | 1 | 1 | 0% | 745 | 732 | -2% | 0 | 0 | — |
case-20 | pass→pass | 3,378 | 1,863 | -45% | 1 | 1 | 0% | 555 | 719 | +30% | 0 | 0 | — |
case-21 | pass→pass | 6,581 | 2,686 | -59% | 1 | 1 | 0% | 1,265 | 955 | -25% | 0 | 0 | — |
case-22 | pass→pass | 3,670 | 3,034 | -17% | 1 | 1 | 0% | 652 | 915 | +40% | 0 | 0 | — |
case-23 | pass→pass | 5,865 | 3,828 | -35% | 1 | 1 | 0% | 1,084 | 1,071 | -1% | 0 | 0 | — |
case-24 | pass→pass | 4,909 | 2,621 | -47% | 1 | 1 | 0% | 927 | 758 | -18% | 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. 24 cases were attempted. The headline lift of +8 percentage points is the difference between those two pass rates over the 24 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.