Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Integrate Supabase Auth with Next.js App Router using middleware, callback handling, protected routes, and server/client client boundaries.
.claude/skills/asymmetric-al-nextjs-supabase-auth/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -11% | 0% |
Use this skill for authentication work in Next.js App Router with Supabase Auth.
Use this skill when:
Do not use this skill when:
supabase-postgres-best-practices)docs/ai/skills/supabase/SKILL.md)@supabase/ssr patterns for App Router integration.@/lib/supabase/server for server-side code and @/lib/supabase/client for client-side code.@/lib/supabase/server@/lib/supabase/clientgetSession blindly in server render paths without user validation| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 16,662 | 12,575 | -25% | 1 | 1 | 0% | 2,990 | 3,032 | +1% | 0 | 0 | — |
case-01 | pass→pass | 14,732 | 11,178 | -24% | 1 | 1 | 0% | 3,081 | 2,697 | -12% | 0 | 0 | — |
case-03 | fail→pass | 12,973 | 8,612 | -34% | 1 | 1 | 0% | 2,450 | 2,138 | -13% | 0 | 0 | — |
case-04 | fail→pass | 7,589 | 2,550 | -66% | 1 | 1 | 0% | 1,316 | 916 | -30% | 0 | 0 | — |
case-05 | pass→pass | 9,835 | 5,522 | -44% | 1 | 1 | 0% | 1,681 | 1,551 | -8% | 0 | 0 | — |
case-06 | pass→pass | 12,556 | 7,290 | -42% | 1 | 1 | 0% | 2,104 | 1,821 | -13% | 0 | 0 | — |
case-11 | fail→pass | 9,558 | 6,359 | -33% | 1 | 1 | 0% | 1,722 | 1,646 | -4% | 0 | 0 | — |
case-07 | fail→pass | 12,303 | 8,490 | -31% | 1 | 1 | 0% | 2,408 | 2,144 | -11% | 0 | 0 | — |
case-08 | pass→pass | 10,932 | 8,025 | -27% | 1 | 1 | 0% | 1,901 | 1,940 | +2% | 0 | 0 | — |
case-09 | pass→pass | 10,864 | 6,360 | -41% | 1 | 1 | 0% | 1,879 | 1,633 | -13% | 0 | 0 | — |
case-10 | pass→pass | 14,464 | 8,216 | -43% | 1 | 1 | 0% | 2,491 | 1,892 | -24% | 0 | 0 | — |
case-12 | fail→pass | 8,659 | 5,256 | -39% | 1 | 1 | 0% | 1,677 | 1,501 | -10% | 0 | 0 | — |
case-13 | fail→pass | 13,770 | 9,196 | -33% | 1 | 1 | 0% | 2,726 | 2,150 | -21% | 0 | 0 | — |
case-14 | fail→pass | 10,929 | 7,126 | -35% | 1 | 1 | 0% | 2,213 | 1,792 | -19% | 0 | 0 | — |
case-15 | fail→pass | 14,209 | 8,353 | -41% | 1 | 1 | 0% | 2,958 | 2,039 | -31% | 0 | 0 | — |
case-16 | fail→pass | 13,468 | 8,101 | -40% | 1 | 1 | 0% | 2,411 | 1,974 | -18% | 0 | 0 | — |
case-17 | pass→pass | 7,336 | 4,559 | -38% | 1 | 1 | 0% | 1,440 | 1,290 | -10% | 0 | 0 | — |
case-18 | pass→pass | 10,503 | 6,030 | -43% | 1 | 1 | 0% | 1,946 | 1,584 | -19% | 0 | 0 | — |
case-19 | fail→pass | 13,965 | 7,347 | -47% | 1 | 1 | 0% | 3,138 | 1,928 | -39% | 0 | 0 | — |
case-20 | pass→pass | 7,895 | 4,868 | -38% | 1 | 1 | 0% | 1,363 | 1,386 | +2% | 0 | 0 | — |
case-21 | pass→pass | 10,539 | 9,866 | -6% | 1 | 1 | 0% | 2,161 | 2,562 | +19% | 0 | 0 | — |
case-22 | pass→pass | 10,200 | 12,300 | +21% | 1 | 1 | 0% | 2,425 | 3,480 | +44% | 0 | 0 | — |
case-23 | fail→pass | 5,752 | 2,937 | -49% | 1 | 1 | 0% | 990 | 1,045 | +6% | 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. 23 cases were attempted. The headline lift of +52 percentage points is the difference between those two pass rates over the 23 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.