Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Implement authentication checks in all public functions
.claude/skills/kunanonj-cursor-plugin-convex-rule-authentication-checks/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-21 | ✓→✓ | = Same ✓ | -20% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 10% | 0% |
Every public function that accesses user data MUST verify authentication using ctx.auth.getUserIdentity().
typescriptexport const getMyTasks = query({ args: {}, handler: async (ctx) => { const identity = await ctx.auth.getUserIdentity(); if (!identity) { throw new Error("Not authenticated"); } const user = await getUserByIdentity(ctx, identity); return await ctx.db .query("tasks") .withIndex("by_user", q => q.eq("userId", user._id)) .collect(); }, });
typescriptexport const updateTask = mutation({ args: { taskId: v.id("tasks"), text: v.string() }, handler: async (ctx, args) => { const identity = await ctx.auth.getUserIdentity(); if (!identity) throw new Error("Not authenticated"); const task = await ctx.db.get(args.taskId); if (!task) throw new Error("Task not found"); const user = await getUserByIdentity(ctx, identity); if (task.userId !== user._id) { throw new Error("Unauthorized"); } await ctx.db.patch(args.taskId, { text: args.text }); }, });
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 9,132 | 6,177 | -32% | 1 | 1 | 0% | 2,182 | 1,736 | -20% | 0 | 0 | — |
case-01 | fail→pass | 5,741 | 4,607 | -20% | 1 | 1 | 0% | 1,310 | 1,403 | +7% | 0 | 0 | — |
case-02 | fail→pass | 8,271 | 5,387 | -35% | 1 | 1 | 0% | 1,828 | 1,568 | -14% | 0 | 0 | — |
case-03 | pass→pass | 6,107 | 4,822 | -21% | 1 | 1 | 0% | 1,344 | 1,480 | +10% | 0 | 0 | — |
case-04 | pass→pass | 5,506 | 5,938 | +8% | 1 | 1 | 0% | 1,319 | 1,814 | +38% | 0 | 0 | — |
case-05 | pass→pass | 6,214 | 4,854 | -22% | 1 | 1 | 0% | 1,407 | 1,530 | +9% | 0 | 0 | — |
case-06 | pass→pass | 8,203 | 6,110 | -26% | 1 | 1 | 0% | 1,731 | 1,914 | +11% | 0 | 0 | — |
case-07 | pass→pass | 8,338 | 6,418 | -23% | 1 | 1 | 0% | 1,985 | 2,006 | +1% | 0 | 0 | — |
case-08 | pass→pass | 6,440 | 4,541 | -29% | 1 | 1 | 0% | 1,482 | 1,446 | -2% | 0 | 0 | — |
case-09 | pass→pass | 7,411 | 4,682 | -37% | 1 | 1 | 0% | 1,705 | 1,509 | -11% | 0 | 0 | — |
case-10 | pass→pass | 6,886 | 4,896 | -29% | 1 | 1 | 0% | 1,497 | 1,490 | -0% | 0 | 0 | — |
case-11 | pass→pass | 6,578 | 5,725 | -13% | 1 | 1 | 0% | 1,643 | 1,904 | +16% | 0 | 0 | — |
case-12 | pass→pass | 4,956 | 5,494 | +11% | 1 | 1 | 0% | 1,106 | 1,559 | +41% | 0 | 0 | — |
case-13 | pass→pass | 5,868 | 5,213 | -11% | 1 | 1 | 0% | 1,321 | 1,681 | +27% | 0 | 0 | — |
case-14 | pass→pass | 5,486 | 5,046 | -8% | 1 | 1 | 0% | 1,213 | 1,521 | +25% | 0 | 0 | — |
case-20 | pass→pass | 10,154 | 5,326 | -48% | 1 | 1 | 0% | 2,055 | 1,458 | -29% | 0 | 0 | — |
case-15 | pass→pass | 4,918 | 4,336 | -12% | 1 | 1 | 0% | 1,077 | 1,439 | +34% | 0 | 0 | — |
case-16 | pass→pass | 7,151 | 4,476 | -37% | 1 | 1 | 0% | 1,530 | 1,511 | -1% | 0 | 0 | — |
case-17 | pass→pass | 8,216 | 5,105 | -38% | 1 | 1 | 0% | 1,958 | 1,629 | -17% | 0 | 0 | — |
case-18 | fail→pass | 5,187 | 5,713 | +10% | 1 | 1 | 0% | 1,191 | 1,765 | +48% | 0 | 0 | — |
case-19 | pass→pass | 7,828 | 4,325 | -45% | 1 | 1 | 0% | 1,724 | 1,355 | -21% | 0 | 0 | — |
case-22 | pass→pass | 7,257 | 4,530 | -38% | 1 | 1 | 0% | 1,602 | 1,442 | -10% | 0 | 0 | — |
case-23 | pass→pass | 2,338 | 1,899 | -19% | 1 | 1 | 0% | 502 | 827 | +65% | 0 | 0 | — |
case-24 | pass→pass | 3,243 | 3,699 | +14% | 1 | 1 | 0% | 734 | 1,034 | +41% | 0 | 0 | — |
case-25 | pass→pass | 5,770 | 4,101 | -29% | 1 | 1 | 0% | 1,340 | 1,319 | -2% | 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 +12 percentage points is the difference between those two pass rates over the 25 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.