Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Firebase gives you a complete backend in minutes - auth, database, storage, functions, hosting. But the ease of setup hides real complexity. Security rules are your last line of defense, and they're often wrong. Firestore queries are limited, and you learn this after you've designed your data model. This skill covers Firebase Authentication, Firestore, Realtime Database, Cloud Functions, Cloud Storage, and Firebase Hosting. Key insight: Firebase is optimized for read-heavy, denormalized data. I
.claude/skills/dokhacgiakhoa-firebase/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 10 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 32% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 76% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 33% | 0% |
You're a developer who has shipped dozens of Firebase projects. You've seen the "easy" path lead to security breaches, runaway costs, and impossible migrations. You know Firebase is powerful, but you also know its sharp edges.
Your hard-won lessons: The team that skipped security rules got pwned. The team that designed Firestore like SQL couldn't query their data. The team that attached listeners to large collections got a $10k bill. You've learned from all of them.
You advocate for Firebase w
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 11,997 | 11,176 | -7% | 1 | 1 | 0% | 1,359 | 1,450 | +7% | 0 | 0 | — |
case-02 | pass→pass | 8,285 | 7,967 | -4% | 1 | 1 | 0% | 765 | 1,011 | +32% | 0 | 0 | — |
case-03 | pass→pass | 8,362 | 9,093 | +9% | 1 | 1 | 0% | 646 | 1,135 | +76% | 0 | 0 | — |
case-04 | pass→pass | 14,220 | 16,078 | +13% | 1 | 1 | 0% | 1,968 | 2,626 | +33% | 0 | 0 | — |
case-05 | pass→pass | 12,632 | 11,897 | -6% | 1 | 1 | 0% | 1,607 | 1,730 | +8% | 0 | 0 | — |
case-06 | pass→pass | 17,154 | 17,528 | +2% | 1 | 1 | 0% | 2,469 | 2,946 | +19% | 0 | 0 | — |
case-07 | pass→pass | 13,861 | 12,622 | -9% | 1 | 1 | 0% | 2,542 | 2,798 | +10% | 0 | 0 | — |
case-08 | pass→pass | 15,654 | 13,482 | -14% | 1 | 1 | 0% | 2,128 | 2,036 | -4% | 0 | 0 | — |
case-09 | fail→pass | 10,894 | 16,460 | +51% | 1 | 1 | 0% | 2,070 | 2,537 | +23% | 0 | 0 | — |
case-10 | pass→pass | 7,184 | 12,065 | +68% | 1 | 1 | 0% | 1,545 | 1,910 | +24% | 0 | 0 | — |
case-11 | pass→pass | 8,978 | 3,892 | -57% | 1 | 1 | 0% | 752 | 1,042 | +39% | 0 | 0 | — |
case-12 | pass→pass | 8,197 | 13,331 | +63% | 1 | 1 | 0% | 1,687 | 2,026 | +20% | 0 | 0 | — |
case-13 | pass→pass | 10,885 | 10,355 | -5% | 1 | 1 | 0% | 2,322 | 2,512 | +8% | 0 | 0 | — |
case-14 | pass→pass | 20,171 | 19,224 | -5% | 1 | 1 | 0% | 2,626 | 2,863 | +9% | 0 | 0 | — |
case-15 | pass→pass | 12,322 | 10,661 | -13% | 1 | 1 | 0% | 1,531 | 1,366 | -11% | 0 | 0 | — |
case-16 | pass→pass | 10,182 | 4,774 | -53% | 1 | 1 | 0% | 941 | 1,190 | +26% | 0 | 0 | — |
case-17 | pass→pass | 12,741 | 15,239 | +20% | 1 | 1 | 0% | 1,590 | 1,885 | +19% | 0 | 0 | — |
case-18 | pass→pass | 11,957 | 8,755 | -27% | 1 | 1 | 0% | 1,395 | 1,998 | +43% | 0 | 0 | — |
case-19 | pass→pass | 12,065 | 12,519 | +4% | 1 | 1 | 0% | 1,461 | 1,775 | +21% | 0 | 0 | — |
case-20 | pass→pass | 14,481 | 6,146 | -58% | 1 | 1 | 0% | 1,890 | 1,511 | -20% | 0 | 0 | — |
case-21 | pass→pass | 11,487 | 14,963 | +30% | 1 | 1 | 0% | 2,103 | 1,906 | -9% | 0 | 0 | — |
case-22 | pass→pass | 13,783 | 12,124 | -12% | 1 | 1 | 0% | 1,747 | 1,694 | -3% | 0 | 0 | — |
case-23 | pass→pass | 11,044 | 10,982 | -1% | 1 | 1 | 0% | 1,089 | 1,500 | +38% | 0 | 0 | — |
case-24 | pass→pass | 6,226 | 12,429 | +100% | 1 | 1 | 0% | 1,346 | 1,723 | +28% | 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 +4 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.