chore: move ai test into example repo

.github/workflows/ai.yml

@@ -21,7 +21,7 @@ jobs:
     env:
       OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
       OPENAI_BASE_URL: ${{ secrets.OPENAI_BASE_URL }}
-      MIDSCENE_MODEL_NAME: gpt-4o-2024-08-06
+      MIDSCENE_MODEL_NAME: gpt-4o-2024-11-20
       CI: 1
       # MIDSCENE_DEBUG_AI_PROFILE: 1
 
@@ -124,7 +124,7 @@ jobs:
         if-no-files-found: ignore
 
     - name: Check if script failed
-      if: steps.test-ai.outcome == 'failure' || steps.e2e-tests.outcome == 'failure' || steps.e2e-tests-cache.outcome == 'failure' || steps.e2e-tests-report.outcome == 'failure'
+      if: steps.test-ai.outcome == 'failure'
       run: exit 1
 
 

README.md

@@ -38,7 +38,7 @@ Midscene.js is an AI-powered automation SDK can control the page, perform assert
 - **Experience by Chrome Extension 🖥️**: Start immediately with the Chrome Extension. No code is needed while exploring.
 - **Visualized Report 🎞️**: With our visualized report file, you can easily understand and debug the whole process.
 - **Out-of-box LLM 🪓**: It is fine to use public multimodal LLMs like GPT-4o. There is no need for any custom training.
-- **Brand New Experience! 🔥**: Experience a whole new world of automation development. Enjoy!
+- **Totally Open Source! 🔥**: Experience a whole new world of automation development. Enjoy!
 
 ## Resources 📄
 
@@ -50,7 +50,7 @@ Midscene.js is an AI-powered automation SDK can control the page, perform assert
   * [Integrate with Puppeteer](https://midscenejs.com/integrate-with-puppeteer.html)
   * [Integrate with Playwright](https://midscenejs.com/integrate-with-playwright.html)
 * [API Reference](https://midscenejs.com/api.html)
-* [Customize Model and Provider](https://midscenejs.com/model-provider.html), see how to use your own model and provider
+* [Customize Model and Provider（e.g. using Gemini）](https://midscenejs.com/model-provider.html), see how to use your own model and provider
 
 ## Community
 

README.zh.md

@@ -35,7 +35,7 @@ Midscene.js 是一个由 AI 驱动的自动化 SDK，能够使用自然语言对
 - **Chrome 插件体验 🖥️**：通过 Chrome 插件，你可以立即开始体验，无需编写代码。
 - **开箱即用的LLM 🪓**：支持使用公开的多模态大语言模型（ 如 GPT-4o ），无需任何定制训练。
 - **可视化报告 🎞️**：通过我们的测试报告和 Playground，你可以轻松理解和调试整个过程。
-- **全新体验 🔥**：体验全新的自动化开发世界，尽情享受吧！
+- **完全开源 🔥**：体验全新的自动化开发世界，尽情享受吧！
 
 ## 资源 📄
 
@@ -47,13 +47,13 @@ Midscene.js 是一个由 AI 驱动的自动化 SDK，能够使用自然语言对
   * [集成到 Puppeteer](https://midscenejs.com/zh/integrate-with-puppeteer.html)
   * [集成到 Playwright](https://midscenejs.com/zh/integrate-with-playwright.html)
 * [API 文档](https://midscenejs.com/zh/api.html)
-* [自定义模型和服务商](https://midscenejs.com/zh/model-provider.html)
+* [自定义模型和服务商（e.g. 使用千问模型）](https://midscenejs.com/zh/model-provider.html)
 
 ## 社区
 
+* [飞书交流群](https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=291q2b25-e913-411a-8c51-191e59aab14d)
 * [Discord](https://discord.gg/AFHJBdnn)
 * [Follow us on X](https://x.com/midscene_ai)
-* [飞书群](https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=291q2b25-e913-411a-8c51-191e59aab14d)
 
   <img src="https://github.com/user-attachments/assets/211b05c9-3ccd-4f52-b798-f3a7f51330ed" alt="lark group link" width="300" />
 

packages/midscene/tests/ai/evaluate/plan/planning.test.ts

@@ -1,7 +1,7 @@
 import { plan } from '@/ai-model';
 /* eslint-disable max-lines-per-function */
 import { describe, expect, it, vi } from 'vitest';
-import { getPageDataOfTestName } from './test-suite/util';
+import { getPageDataOfTestName } from '../test-suite/util';
 
 vi.setConfig({
   testTimeout: 180 * 1000,

packages/midscene/tests/ai/plan/plan-to-target.test.ts

@@ -1,6 +1,5 @@
 import path from 'node:path';
-import { AiInspectElement } from '@/ai-model';
-import { vlmPlanning } from '@/ai-model/vlm-planning';
+import { vlmPlanning } from '@/ai-model/ui-tars-planning';
 import { savePositionImg } from '@midscene/shared/img';
 import { expect, test } from 'vitest';
 import { getPageTestData } from '../evaluate/test-suite/util';

packages/midscene/tests/ai/prompt.test.ts

@@ -1,8 +1,5 @@
-import path from 'node:path';
-import { AiInspectElement } from '@/ai-model';
-import { systemPromptToTaskPlanning } from '@/ai-model/prompt/planning';
+import { systemPromptToTaskPlanning } from '@/ai-model/prompt/llm-planning';
 import { describe, expect, it, test } from 'vitest';
-import { getPageTestData } from './evaluate/test-suite/util';
 
 describe('automation - computer', () => {
   it('should be able to generate prompt', async () => {