Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Point a domain you already own at your Convex app (DNS records, custom-domain attach, auth-origin rebind).
.claude/skills/get-convex-convex-domains/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-24 | ✗→✓ | ▲ Improved | 61% | 0% |
<!-- GENERATED from convex-agents content/capabilities/domains.json — do not edit by hand. -->
Walk the user's own registrar through pointing their domain at the Convex app: identify the target (hosting or deployment URL), create the DNS records, attach the custom domain, and rebind the auth origin if the app uses auth.
*.convex.app static hosting) or the deployment's HTTP actions URL.flarectl dns create (note: wrangler itself doesn't manage DNS records) or the CF API via their token env; Route53 → aws route53 change-resource-record-sets; Google Cloud DNS → gcloud dns record-sets create; DigitalOcean → doctl compute domain records create; Vercel DNS → vercel dns add. Check auth read-only first (flarectl user info / aws sts get-caller-identity / doctl account get); show the exact commands and get a yes before running.dig +short.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→fail | 11,105 | 10,419 | -6% | 1 | 1 | 0% | 1,924 | 2,419 | +26% | 0 | 0 | — |
case-02 | fail→fail | 13,116 | 7,796 | -41% | 1 | 1 | 0% | 2,317 | 1,830 | -21% | 0 | 0 | — |
case-01 | fail→fail | 18,328 | 15,351 | -16% | 1 | 1 | 0% | 2,605 | 3,237 | +24% | 0 | 0 | — |
case-03 | fail→fail | 15,871 | 12,315 | -22% | 1 | 1 | 0% | 2,858 | 1,385 | -52% | 0 | 0 | — |
case-04 | pass→pass | 13,505 | 7,484 | -45% | 1 | 1 | 0% | 1,890 | 1,802 | -5% | 0 | 0 | — |
case-05 | pass→pass | 18,036 | 18,537 | +3% | 1 | 1 | 0% | 3,188 | 2,293 | -28% | 0 | 0 | — |
case-06 | pass→pass | 11,568 | 12,612 | +9% | 1 | 1 | 0% | 2,143 | 2,415 | +13% | 0 | 0 | — |
case-07 | fail→fail | 13,653 | 9,164 | -33% | 1 | 1 | 0% | 1,899 | 2,256 | +19% | 0 | 0 | — |
case-09 | pass→pass | 11,998 | 8,484 | -29% | 1 | 1 | 0% | 1,780 | 1,787 | +0% | 0 | 0 | — |
case-10 | pass→pass | 11,224 | 8,486 | -24% | 1 | 1 | 0% | 1,857 | 1,870 | +1% | 0 | 0 | — |
case-11 | fail→pass | 11,024 | 6,843 | -38% | 1 | 1 | 0% | 1,575 | 1,501 | -5% | 0 | 0 | — |
case-12 | fail→pass | 13,347 | 11,211 | -16% | 1 | 1 | 0% | 1,939 | 2,581 | +33% | 0 | 0 | — |
case-13 | fail→fail | 13,085 | 10,215 | -22% | 1 | 1 | 0% | 1,850 | 2,429 | +31% | 0 | 0 | — |
case-14 | pass→pass | 10,854 | 8,432 | -22% | 1 | 1 | 0% | 1,454 | 2,114 | +45% | 0 | 0 | — |
case-15 | fail→pass | 10,850 | 12,669 | +17% | 1 | 1 | 0% | 1,577 | 2,409 | +53% | 0 | 0 | — |
case-16 | pass→pass | 12,727 | 9,609 | -24% | 1 | 1 | 0% | 2,257 | 2,158 | -4% | 0 | 0 | — |
case-17 | pass→pass | 10,301 | 9,061 | -12% | 1 | 1 | 0% | 1,458 | 2,177 | +49% | 0 | 0 | — |
case-18 | pass→pass | 11,362 | 11,196 | -1% | 1 | 1 | 0% | 1,955 | 2,063 | +6% | 0 | 0 | — |
case-19 | fail→fail | 10,188 | 7,412 | -27% | 1 | 1 | 0% | 1,540 | 1,566 | +2% | 0 | 0 | — |
case-20 | pass→pass | 5,663 | 3,727 | -34% | 1 | 1 | 0% | 710 | 1,025 | +44% | 0 | 0 | — |
case-21 | pass→pass | 11,006 | 8,269 | -25% | 1 | 1 | 0% | 1,538 | 1,755 | +14% | 0 | 0 | — |
case-22 | fail→pass | 13,656 | 9,224 | -32% | 1 | 1 | 0% | 1,869 | 2,062 | +10% | 0 | 0 | — |
case-23 | pass→pass | 15,310 | 8,761 | -43% | 1 | 1 | 0% | 2,395 | 1,976 | -17% | 0 | 0 | — |
case-24 | fail→pass | 4,999 | 6,709 | +34% | 1 | 1 | 0% | 938 | 1,513 | +61% | 0 | 0 | — |
case-25 | pass→pass | 9,109 | 3,765 | -59% | 1 | 1 | 0% | 1,549 | 1,227 | -21% | 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. 25 cases were attempted. The headline lift of +20 percentage points is the difference between those two pass rates over the 25 comparable cases. 1 case got worse with the skill loaded, and it is 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.