docs: add lost doc

Author: phodal

docs/agent/custom-ai-agent.md

@@ -28,16 +28,19 @@ Demo:
 [
   {
     "name": "内部 API 集成",
+    "description": "在一个组织或项目中，不同系统或组件之间的通信接口。",
     "url": "http://127.0.0.1:8765/api/agent/api-market",
     "responseAction": "Direct"
   },
   {
     "name": "组件库查询",
+    "description": "从组件库中检索特定的 UI 组件，以便在开发的应用程序中使用。",
     "url": "http://127.0.0.1:8765/api/agent/component-list",
     "responseAction": "TextChunk"
   },
   {
     "name": "页面生成",
+    "description": "使用 React 框架，基于组件和状态来生成页面。",
     "url": "http://127.0.0.1:8765/api/agent/ux",
     "auth": {
       "type": "Bearer",

docs/patterns/by-example.md

@@ -5,23 +5,23 @@ nav_order: 1
 parent: Design Patterns
 ---
 
-根据当前的用户习惯，寻找待生成的内容作为示例，而后作为生成式 AI 的输入。
+> Finding example content as context to generate as input for the generative AI, based on current user habits.
 
-## 提交信息示例
+## Git Commit message Example
 
-步骤：
+Implementation steps:
 
-1. 获取当前项目的版本控制系统（VCS）日志提供者。
-2. 获取当前分支和用户。
-3. 根据用户或分支过滤日志。
-4. 收集示例提交信息。
+1. Retrieve version control system (VCS) log providers for the current project.
+2. Get the current branch and user.
+3. Filter logs based on user or branch.
+4. Collect example submission information.
 
-代码示例：
+### Code Implementation
 
 ```kotlin
 private fun findExampleCommitMessages(project: Project): String? {
     val logProviders = VcsProjectLog.getLogProviders(project)
-    val entry = logProviders.entries.firstOrNull() ?: return null
+    logProviders.entries.firstOrNull() ?: return null
 
     val logProvider = entry.value
     val branch = logProvider.getCurrentBranch(entry.key) ?: return null
@@ -37,18 +37,43 @@ private fun findExampleCommitMessages(project: Project): String? {
 }
 ```
 
-## 代码示例：测试代码示例
+### Template Example
+       
+    Write a cohesive yet descriptive commit message for a given diff.
+    - Make sure to include both information What was changed and Why.
+    - Start with a short sentence in imperative form, no more than 50 characters long.
+    - Then leave an empty line and continue with a more detailed explanation, if necessary.
+    - Explanation should have less than 200 characters.
+    
+    Follow the Conventional Commits specification, examples:
+    - fix(authentication): fix password regex pattern case
+    - feat(storage): add support for S3 storage
+    - test(java): fix test case for user controller
+    - docs(docs): add architecture diagram to home page
+    
+    #if( $context.historyExamples.length() > 0 )
+    Here is History Examples:
+    $context.historyExamples
+    #end
+    
+    Diff:
+    
+    ```diff
+    ${context.diffContent}
 
-参考语言实现：`JavaTestContextProvider`、`KotlinTestContextProvider`。
+## Code Example: Testing Code Samples
 
-步骤：
+Reference language implementations: `JavaTestContextProvider`, `KotlinTestContextProvider`.
 
-1. 获取当前项目的被测试代码。
-2. 根据被测试代码寻找模板：`Controller`、`Service`、默认测试等。
-    - Kotlin：`ControllerTest.kt`, `ServiceTest.kt`, `Test.kt`
-    - Java: `ControllerTest.java`, `ServiceTest.java`, `Test.java`
-3. 根据模板生成测试代码。
+Steps:
 
-## 文档示例
+1. Retrieve the tested code for the current project.
+2. Find templates based on the tested code: `Controller`, `Service`, default tests, etc.
+   - Kotlin: `ControllerTest.kt`, `ServiceTest.kt`, `Test.kt`
+   - Java: `ControllerTest.java`, `ServiceTest.java`, `Test.java`
+3. Generate testing code based on templates.
+
+## Document Example
 
 DOC TODO
+```

docs/patterns/task-driven.md

@@ -5,4 +5,11 @@ nav_order: 3
 parent: Design Patterns
 ---
 
-// TODO
+In issue [#79](https://github.com/unit-mesh/auto-dev/issues/79), we defined the concept of "task-driven AI generation"
+as a method or concept that emphasizes the use of artificial intelligence (AI) generation targeted at specific tasks or
+goals to meet particular needs or goals. In this context, tasks can refer to various modules in software development,
+such as function blocks, class blocks, test methods, SQL snippets, code snippets, or even user requirements, among
+others.
+
+## Task Driven
+

docs/patterns/unit-driven.md

@@ -5,15 +5,33 @@ nav_order: 2
 parent: Design Patterns
 ---
 
-"单元驱动 AI 生成" 是一种方法或理念，它强调通过针对具体的单元或模块进行人工智能（AI）生成，以满足特定的需求或目标。在这个上下文中，
-单元可以是各种软件开发中的模块，如函数块、类块、测试方法、SQL 片段、代码片段，甚至可以是用户需求等等。
+> "Unit-driven AI generation" is a method or concept that emphasizes the use of artificial intelligence (AI) generation
+> targeted at specific units or modules to meet particular needs or goals. In this context, units can refer to various
+> modules in software development, such as function blocks, class blocks, test methods, SQL snippets, code snippets, or
+> even user requirements, among others.
 
-这种方法的核心思想是将整个系统或应用划分为小的、独立的单元，然后使用AI技术为每个单元生成相应的代码、脚本、查询或其他必要的实现。
-这种方法有助于提高开发效率，减少重复工作，同时确保每个单元都能够独立运作，并且在整个系统中具有一致性。
+The core idea of this approach is to divide the entire system or application into small, independent units and then use
+AI technology to generate corresponding code, scripts, queries, or other necessary implementations for each unit. This
+approach helps improve development efficiency, reduce repetitive work, and ensure that each unit can operate
+independently and consistently throughout the system.
 
-具体而言，单元驱动 AI 生成可以涉及到使用自然语言处理（NLP）技术解析用户需求，然后生成对应的代码片段；或者针对特定的编程语言、数据库查询语言等，
-通过学习和推理生成相应的单元实现。这种方法的优势在于能够更灵活地适应不同的需求和环境，同时提高开发的速度和质量。
+Specifically, unit-driven AI generation may involve using natural language processing (NLP) techniques to parse user
+requirements and then generate corresponding code snippets. Alternatively, it could involve learning and inference to
+generate corresponding unit implementations for specific programming languages, database query languages, etc. The
+advantage of this approach is its flexibility in adapting to different requirements and environments while increasing
+the speed and quality of development.
 
-## Unit Driven 示例
+## Unit Driven Example
+
+### AutoSQL
+
+see in [AutoSQL](https://github.com/unit-mesh/auto-dev/tree/master/exts/database/src/main/kotlin/cc/unitmesh/database/flow)
+
+### AutoPage
+
+see in [AutoPage](https://github.com/unit-mesh/auto-dev/tree/master/javascript/src/main/kotlin/cc/unitmesh/ide/javascript/flow)
+
+### AutoArkUI
+
+see in [AutoArkUI](https://github.com/unit-mesh/auto-dev/tree/master/exts/ext-harmonyos/src/main/kotlin/cc/unitmesh/harmonyos/actions/auto)
 
-AutoSQL,AutoPage,AutoTest