Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Skill for working with Firebase Hosting (Classic). Use this when you want to deploy static web apps, Single Page Apps (SPAs), or simple microservices. Do NOT use for Firebase App Hosting.
.claude/skills/firebase-hosting-basics/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -22% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -32% | 0% |
This skill provides instructions and references for working with Firebase Hosting, a fast and secure hosting service for your web app, static and dynamic content, and microservices.
Firebase Hosting provides production-grade web content hosting for developers. With a single command, you can deploy web apps and serve both static and dynamic content to a global CDN (content delivery network).
Key Features:
world.
deploying live.
Functions or Cloud Run.
Choose Firebase Hosting if:
Choose Firebase App Hosting if:
firebase.json)For details on configuring Hosting behavior, including public directories, redirects, rewrites, and headers, see configuration.md.
For instructions on deploying your site, using preview channels, and managing releases, see deploying.md.
To test your app locally:
bashnpx -y firebase-tools@latest emulators:start --only hosting
This serves your app at http://localhost:5000 by default.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 7,645 | 1,919 | -75% | 1 | 1 | 0% | 1,567 | 762 | -51% | 0 | 0 | — |
case-02 | fail→pass | 12,368 | 8,733 | -29% | 1 | 1 | 0% | 2,538 | 2,347 | -8% | 0 | 0 | — |
case-03 | pass→pass | 9,794 | 4,575 | -53% | 1 | 1 | 0% | 1,583 | 1,238 | -22% | 0 | 0 | — |
case-04 | pass→pass | 10,207 | 4,381 | -57% | 1 | 1 | 0% | 1,760 | 1,201 | -32% | 0 | 0 | — |
case-05 | pass→pass | 6,213 | 2,434 | -61% | 1 | 1 | 0% | 1,111 | 860 | -23% | 0 | 0 | — |
case-06 | pass→pass | 9,478 | 2,832 | -70% | 1 | 1 | 0% | 1,540 | 936 | -39% | 0 | 0 | — |
case-07 | pass→pass | 9,831 | 7,233 | -26% | 1 | 1 | 0% | 1,809 | 1,747 | -3% | 0 | 0 | — |
case-08 | pass→pass | 7,367 | 2,050 | -72% | 1 | 1 | 0% | 1,264 | 708 | -44% | 0 | 0 | — |
case-09 | pass→pass | 4,518 | 3,408 | -25% | 1 | 1 | 0% | 670 | 986 | +47% | 0 | 0 | — |
case-10 | pass→pass | 9,992 | 5,357 | -46% | 1 | 1 | 0% | 1,748 | 1,408 | -19% | 0 | 0 | — |
case-11 | pass→pass | 11,842 | 5,051 | -57% | 1 | 1 | 0% | 1,970 | 1,387 | -30% | 0 | 0 | — |
case-12 | pass→pass | 7,344 | 1,459 | -80% | 1 | 1 | 0% | 1,326 | 654 | -51% | 0 | 0 | — |
case-13 | fail→pass | 9,989 | 3,382 | -66% | 1 | 1 | 0% | 1,868 | 1,029 | -45% | 0 | 0 | — |
case-14 | pass→pass | 3,458 | 1,589 | -54% | 1 | 1 | 0% | 599 | 665 | +11% | 0 | 0 | — |
case-15 | pass→pass | 7,533 | 4,748 | -37% | 1 | 1 | 0% | 1,447 | 1,361 | -6% | 0 | 0 | — |
case-16 | pass→pass | 4,424 | 2,670 | -40% | 1 | 1 | 0% | 838 | 852 | +2% | 0 | 0 | — |
case-17 | pass→pass | 6,223 | 4,693 | -25% | 1 | 1 | 0% | 1,180 | 1,311 | +11% | 0 | 0 | — |
case-18 | pass→pass | 8,368 | 5,190 | -38% | 1 | 1 | 0% | 1,597 | 1,381 | -14% | 0 | 0 | — |
case-19 | pass→pass | 6,393 | 3,423 | -46% | 1 | 1 | 0% | 1,255 | 1,047 | -17% | 0 | 0 | — |
case-20 | pass→pass | 2,788 | 2,533 | -9% | 1 | 1 | 0% | 402 | 823 | +105% | 0 | 0 | — |
case-21 | pass→pass | 2,616 | 2,596 | -1% | 1 | 1 | 0% | 428 | 830 | +94% | 0 | 0 | — |
case-22 | pass→pass | 3,926 | 2,707 | -31% | 1 | 1 | 0% | 601 | 882 | +47% | 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 +14 percentage points is the difference between those two pass rates over the 22 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/24/2026 | +32% |
Other measured skills in the registry, with their headline benchmark lift.