Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Open the custom UI for a workflow run in your browser. Starts a local Gateway automatically if none is running (serving workflow-owned <UI> declarations); pass --no-autostart or --gateway <url> to opt out. Run `smithers ui --help` for usage details.
.claude/skills/smithersai-smithers-ui/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -73% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -60% | 0% |
Open the custom UI for a workflow run in your browser. Starts a local Gateway automatically if none is running (serving workflow-owned <UI> declarations); pass --no-autostart or --gateway <url> to opt out.
| Name | Type | Required | Description | |------|------|----------|-------------| | runId | string | no | Run to open. Defaults to the most recent run. |
| Flag | Type | Default | Description | |------|------|---------|-------------| | --gateway | string | | Gateway base URL (default http://127.0.0.1:<port>). | | --port | number | 7331 | Gateway port when --gateway is not set. | | --workflow | string | | Open this workflow's UI directly, skipping run lookup. | | --app | boolean | false | Open the full local Smithers UI (the apps/smithers control surface) instead of a single workflow run UI. Builds the bundle on first use and serves it against the local Gateway. | | --appPort | number | 7332 | Port to serve the full UI on (with --app). | | --rebuild | boolean | false | Force a rebuild of the full UI bundle before serving (with --app). | | --open | boolean | true | Open a browser. Use --no-open to just print the URL. | | --autostart | boolean | true | If no Gateway is reachable on the local port, start one automatically. Use --no-autostart to disable. | | --daemon | boolean | true | Allow a background gateway daemon. Use --no-daemon (or SMITHERS_NO_DAEMON=1) to force direct/embedded operation and never autostart one — for CI, sandboxes, and containers. |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | fail→fail | 13,468 | 5,892 | -56% | 1 | 1 | 0% | 2,320 | 1,472 | -37% | 0 | 0 | — |
case-01 | fail→fail | 7,671 | 7,523 | -2% | 1 | 1 | 0% | 929 | 693 | -25% | 0 | 0 | — |
case-02 | fail→pass | 8,679 | 2,626 | -70% | 1 | 1 | 0% | 1,331 | 822 | -38% | 0 | 0 | — |
case-03 | fail→pass | 17,912 | 7,963 | -56% | 1 | 1 | 0% | 3,460 | 935 | -73% | 0 | 0 | — |
case-04 | fail→pass | 6,867 | 3,478 | -49% | 1 | 1 | 0% | 1,242 | 1,092 | -12% | 0 | 0 | — |
case-05 | fail→pass | 7,626 | 2,324 | -70% | 1 | 1 | 0% | 1,274 | 790 | -38% | 0 | 0 | — |
case-06 | fail→pass | 10,854 | 2,299 | -79% | 1 | 1 | 0% | 2,040 | 819 | -60% | 0 | 0 | — |
case-07 | fail→fail | 6,200 | 3,290 | -47% | 1 | 1 | 0% | 1,080 | 653 | -40% | 0 | 0 | — |
case-08 | fail→pass | 3,944 | 2,380 | -40% | 1 | 1 | 0% | 633 | 809 | +28% | 0 | 0 | — |
case-09 | fail→pass | 5,548 | 3,998 | -28% | 1 | 1 | 0% | 835 | 1,056 | +26% | 0 | 0 | — |
case-10 | fail→pass | 33,906 | 1,575 | -95% | 1 | 1 | 0% | 6,163 | 678 | -89% | 0 | 0 | — |
case-21 | fail→pass | 13,754 | 9,611 | -30% | 1 | 1 | 0% | 2,568 | 2,340 | -9% | 0 | 0 | — |
case-11 | fail→pass | 11,775 | 1,450 | -88% | 1 | 1 | 0% | 2,028 | 611 | -70% | 0 | 0 | — |
case-12 | fail→pass | 11,642 | 1,106 | -90% | 1 | 1 | 0% | 2,042 | 576 | -72% | 0 | 0 | — |
case-13 | pass→pass | 3,498 | 1,142 | -67% | 1 | 1 | 0% | 553 | 590 | +7% | 0 | 0 | — |
case-14 | pass→pass | 4,337 | 5,442 | +25% | 1 | 1 | 0% | 717 | 1,296 | +81% | 0 | 0 | — |
case-15 | fail→pass | 14,724 | 1,834 | -88% | 1 | 1 | 0% | 2,228 | 686 | -69% | 0 | 0 | — |
case-16 | fail→pass | 6,057 | 1,221 | -80% | 1 | 1 | 0% | 927 | 613 | -34% | 0 | 0 | — |
case-17 | pass→pass | 8,865 | 1,440 | -84% | 1 | 1 | 0% | 1,404 | 621 | -56% | 0 | 0 | — |
case-18 | pass→pass | 11,754 | 1,320 | -89% | 1 | 1 | 0% | 1,971 | 635 | -68% | 0 | 0 | — |
case-19 | fail→fail | 5,180 | 1,361 | -74% | 1 | 1 | 0% | 761 | 680 | -11% | 0 | 0 | — |
case-20 | fail→fail | 6,188 | 2,982 | -52% | 1 | 1 | 0% | 1,071 | 919 | -14% | 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 20 counted toward the lift figure. The other 2 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 +59 percentage points is the difference between those two pass rates over the 20 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.