prototype to show how we can validate token names

src/schemas/tokenName.ts

@@ -1,12 +1,28 @@
 import {z} from 'zod'
 import {schemaErrorMessage} from '../utilities/schemaErrorMessage'
 
-export const tokenName = z.string().refine(
-  name => /(^[a-z0-9][A-Za-z0-9-]*$|^@$)/.test(name),
-  name => ({
-    message: schemaErrorMessage(
-      `Invalid token name: "${name}"`,
-      'Token name must be kebab-case or camelCase, and start with a lowercase letter or number and consist only of letters, numbers, and hyphens.',
-    ),
-  }),
-)
+export const tokenName = z
+  .string()
+  .refine(
+    name => /(^[a-z0-9][A-Za-z0-9-]*$|^@$)/.test(name),
+    name => ({
+      message: schemaErrorMessage(
+        `Invalid token name: "${name}"`,
+        'Token name must be kebab-case or camelCase, and start with a lowercase letter or number and consist only of letters, numbers, and hyphens.',
+      ),
+    }),
+  )
+  .superRefine((val, ctx) => {
+    // validate base tokens
+    const name = ctx.path.join('-')
+    if (ctx.path[0] === 'base' && /(^base-[text|color|size]-[A-Za-z0-9-]*$)/.test(name)) {
+      ctx.addIssue({
+        code: z.ZodIssueCode.invalid_string,
+        validation: 'regex',
+        message: schemaErrorMessage(
+          `Invalid base token name: "${name}"`,
+          'Base token names must be kebab-case or camelCase, and start with a lowercase letter or number and consist only of letters, numbers, and hyphens.',
+        ),
+      })
+    }
+  })