Working on it

jest.config.cjs

@@ -0,0 +1,17 @@
+/** @type {import('ts-jest').JestConfigWithTsJest} */
+module.exports = {
+  testEnvironment: "node",
+  transform: {
+    "\\.yaml$": "jest-text-transformer",
+    "\\.[jt]sx?$": "ts-jest",
+  },
+  globals: {
+    "ts-jest": {
+      useESM: true,
+    },
+  },
+  moduleNameMapper: {
+    "(.+)\\.js": "$1",
+  },
+  extensionsToTreatAsEsm: [".ts"],
+};

node/llm.test.ts

@@ -1,5 +1,3 @@
-import path from "path";
-
 import { Battle, LLMClassifierFromTemplate, OpenAIClassifier } from "../js/llm";
 import { ChatCompletionRequestMessage } from "openai";
 import { ChatCache } from "../js/oai";
@@ -59,7 +57,7 @@ Nicolo also dropped this as a reference: http://spec.openapis.org/oas/v3.0.3#ope
 
   expect(score.score).toBe(1);
   expect(score.error).toBeUndefined();
-}, 60000);
+}, 600000);
 
 test("llm_classifier", async () => {
   const pageContent = `As suggested by Nicolo, we should standardize the error responses coming from GoTrue, postgres, and realtime (and any other/future APIs) so that it's better DX when writing a client,
@@ -106,7 +104,7 @@ Issue Description: {{page_content}}
     expect(response.score).toBe(0);
     expect(response.error).toBeUndefined();
   }
-}, 60000);
+}, 600000);
 
 test("battle", async () => {
   for (const useCoT of [true, false]) {
@@ -144,4 +142,4 @@ test("battle", async () => {
     expect(response.score).toBe(0);
     expect(response.error).toBeUndefined();
   }
-}, 60000);
+}, 600000);