@buaa_smat/hometrans 0.1.15 → 0.1.16

package/README.md

@@ -2,32 +2,31 @@
 
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![Node.js](https://img.shields.io/badge/Node.js-%3E%3D%2022-339933.svg?logo=node.js&logoColor=white)](https://nodejs.org/en/download)
+
+[![Claude Code](https://img.shields.io/badge/Claude_Code-supported-blueviolet.svg)](#选择本地editor)
+[![Cursor](https://img.shields.io/badge/Cursor-supported-blueviolet.svg)](#选择本地editor)
+[![OpenCode](https://img.shields.io/badge/OpenCode-supported-blueviolet.svg)](#选择本地editor)
+[![DevEco Code](https://img.shields.io/badge/DevEco_Code-supported-blueviolet.svg)](#选择本地editor)
+[![Codex](https://img.shields.io/badge/Codex-supported-blueviolet.svg)](#选择本地editor)
+[![CodeBuddy](https://img.shields.io/badge/CodeBuddy-supported-blueviolet.svg)](#选择本地editor)
 
 本文档为完整的Android到HarmonyOS迁移流程指导。
 
 ---
 
-## 步骤0:安装依赖
-
-### 0.1安装依赖并连接设备
-
-按以下顺序安装(`ht init` 会逐项检测,并提示缺失项对各skill的影响)。各依赖下方均列出了「被依赖」的步骤:如果你不需要执行某依赖所对应的步骤,则可以不安装该依赖(相应skill会显示为受限或不可用,但不影响其它步骤)。
+## 安装DevEco Studio以及Android Studio
 
-1. **安装DevEco Studio** — 下载地址:<https://developer.huawei.com/consumer/cn/download/>。安装完成后,配置以下 **PATH环境变量**(`<DevEco安装目录>` 指DevEco Studio的安装根目录):
+1. **安装DevEco Studio** — 下载地址:<https://developer.huawei.com/consumer/cn/download/>。安装完成后,配置以下 **PATH环境变量**:
 
    | 工具 | PATH中加入的目录 |
       |------|------------------|
-   | node / npm / npx | `<DevEco安装目录>\tools\node` |
    | java | `<DevEco安装目录>\jbr\bin` |
    | hdc | `<DevEco安装目录>\sdk\default\openharmony\toolchains` |
 
-   验证方式:执行 `node -v`、`java -version`、`hdc -v` 均返回正常
+   验证方式:执行 `java -version`、`hdc -v` 均返回正常
+
 
-   被依赖:
-   - 资源转换skill(java可选,缺失时回退读取源码 `res/`)
-   - 步骤1.2UI对齐(必需)
-   - 步骤3逻辑代码转换流水线(必需,构建/评审修复)
-   - 集成测试skill(必需,真机调试)
 2. **安装Android Studio** — 下载地址:<https://developer.android.com/studio>,并在 **SDK Manager** 中安装SDK。
 
    验证方式:执行 `adb version` 返回正常
@@ -35,96 +34,62 @@
    被依赖:
    - 步骤1.1UI迁移(可选,仅自动抓取页面快照时需要)
    - 步骤1.2UI对齐(必需)
-3. **安装uv** — 安装指引:<https://docs.astral.sh/uv/getting-started/installation/>。
 
-   验证方式:执行 `uv  --version` 返回正常
+## 安装Node.js
 
-   被依赖:
-   - 步骤3逻辑代码转换流水线的自测阶段(可经 `skip_test` 跳过)
-   - 集成测试skill(必需,AutoTest在uv下运行)
-4. **安装python** — 安装指引:<https://www.python.org/downloads/>,要求 ≥ 3.10。
+安装 **Node.js(版本需 >= 22)** — 下载地址:<https://nodejs.org/en/download>。
 
-   验证方式:执行 `python  --version` 返回正常
+验证方式:执行 `node -v`、`npm -v`、`npx -v` 均返回正常,且 `node -v` 显示版本不低于 22
 
-   被依赖:
-   - 步骤1.1UI迁移(必需,页面快照解析脚本)
-   - 步骤1.2UI对齐(必需,双端页面采集脚本)
-5. **安装 [GitNexus](https://github.com/abhigyanpatwari/GitNexus)** — 执行 `npm install -g gitnexus && gitnexus setup`
+## 安装HomeTrans
 
-   验证方式:执行 `gitnexus --version` 返回正常
+执行 `npm install -g @buaa_smat/hometrans`。
 
-   被依赖:
-   - 步骤2生成需求规格(必需)
-6. **连接设备** — 连接安卓和鸿蒙真机,或启动模拟器
+验证方式:执行 `hometrans --version` 或 `ht --version` 返回正常
 
-   被依赖:
-   - 安卓真机/模拟器 — 步骤1 UI迁移
-   - 鸿蒙真机/模拟器 — 步骤1.2UI对齐
-   - 鸿蒙**真机** — 步骤3流水线自测阶段、集成测试skill
+## 环境初始化
 
-### 0.2安装hometrans
+执行 `hometrans init` 或 `ht init`。
 
-执行 `npm install -g @buaa_smat/hometrans`。
+> 如果在PowerShell下运行,命令需要加上 `.cmd` 后缀,如 `hometrans.cmd --version`、`ht.cmd init`。
 
-验证方式:执行 `hometrans --version` 或 `ht --version` 返回正常
+### 选择本地editor:
+![choose_editor.png](https://raw.gitcode.com/SMAT/HomeTrans/raw/main/resource/choose_editor.png)
 
-然后执行 `hometrans init` 或 `ht init`。
+### 参数配置:
 
+#### 配置鸿蒙SDK:
+![set_sdk.png](https://raw.gitcode.com/SMAT/HomeTrans/raw/main/resource/set_sdk.png)
 
-> 如果在PowerShell下运行,命令需要加上 `.cmd` 后缀,如 `hometrans.cmd --version`、`ht.cmd init`。
+#### 配置多模态模型(不执行UI对齐以及集成测试可以跳过该配置):
+![set_multimodel.png](https://raw.gitcode.com/SMAT/HomeTrans/raw/main/resource/set_multimodel.png)
+
+  | 参数             | 含义             |
+  |-----------------|---------|
+  | `model_api_key` | 用于模型API请求认证的令牌   |
+  | `model_name`      | 多模态(视觉)模型的名字   |
+  | `model_base_url`       | 模型API服务的基础地址 |
+
+##### 配置参考
+
+前往[阿里云百炼平台](https://bailian.console.aliyun.com/)开通服务并申请 API Key,选用 Qwen 系列多模态(视觉)模型(如 `qwen3.5-plus`)。
+
+| 参数 | 取值                                                  |
+|---|-----------------------------------------------------|
+| `model_api_key` | 百炼控制台申请的 API Key(形如 `sk-xxxxxxxx`)                  |
+| `model_name` | `qwen3.5-plus`                                      |
+| `model_base_url` | `https://dashscope.aliyuncs.com/compatible-mode/v1` |
+
+#### 将配置信息添加到环境变量:
+![set_environment.png](https://raw.gitcode.com/SMAT/HomeTrans/raw/main/resource/set_environment.png)
 
-### 0.2.1选择本地editor:
-![choose_editor.png](resource/choose_editor.png)
-
-### 0.2.2通用配置:
-![common_config.png](resource/common_config.png)
-- `DEVECO_SDK_HOME` — DevEco Studio的sdk目录,其余路径由它派生并校验存在。示例:
-
-  ```text
-  + DEVECO_SDK_HOME : <DevEco安装目录>\sdk
-  + DEVECO_HOME     : <DevEco安装目录>
-  + OHOS_SDK_PATH   : <DevEco安装目录>\sdk\default\openharmony\ets
-  + HMS_SDK_PATH    : <DevEco安装目录>\sdk\default\hms\ets
-  ```
-### 0.2.3 UI对齐配置(如果不执行步骤1.2UI对齐,则可以不配置):
-![ui_align_config.png](resource/ui_align_config.png)
-- `GLM_API_KEY` — UI对齐中过程中GLM phone agent所用的LLM API key。
-
-  申请方式：在[智谱官方网站](https://bigmodel.cn/apikey/platform)注册用户后，申请一个API KEY即可，调用的是免费的auto-glm模型，无需充值。
-
-### 0.2.4集成测试配置(如果不执行步骤3的自测阶段与集成测试skill,则可以不配置):
-![integration_test_config.png](resource/integration_test_config.png)
-- `autotest api_key` — 集成测试agent执行测试用例(真机自测)所用的多模态大模型API key。各配置项含义如下,`init` 环节只配置前4项,其余参数在`config.json`中配置。
-
-  | 参数                | 示例 / 默认值 | 含义 |
-  |-------------------|--------------|------|
-  | `name`            | `""` | 多模态(视觉)模型的名字 |
-  | `base_url`        | `""` | 模型API服务的基础地址 |
-  | `api_key`         | (需填写) | 用于模型API请求认证的令牌 |
-  | `provider`        | `openai` | 模型API服务的提供方标识,如openai;Azure服务填azure |
-  | `temperature`     | `0.1` | 采样温度 |
-  | `top_p`           | `0.9` | 核采样阈值 |
-  | `frequency_penalty` | `0.5` | 重复惩罚 |
-  | `device_sn`       | `null` | 设备序列号;`null` 表示自动选用当前连接的设备 |
-  | `ip`              | `127.0.0.1` | 设备连接IP(hdc) |
-  | `port`            | `8710` | 设备连接端口(hdc) |
-  | `device_log_level` | `debug` | 设备日志级别:debug/info/... |
-  | `max_steps`       | `50` | 单条用例最大执行步数 |
-  | `verbose`         | `true` | 是否输出详细日志 |
-  | `lang`            | `zh` | 输出语言:zh/en |
-  | `enable_preprocess` | `false` | 执行前是否先用unified_model把用例拆成步骤序列(多一次LLM调用,默认关) |
-  | `debug_mode`      | `null` | 调试会话:`null` 关闭 / `memory` 内存 / `file` 存盘并跑完后起调试服务(:5000) |
-
-### 0.2.5将配置信息添加到环境变量:
-![set_env.png](resource/set_env.png)
-
-### 0.3准备项目
+### 准备项目
 
 准备 **Android源项目** 与 **HarmonyOS目标项目目录**。
 
 ---
 ## 迁移流程
-![migration_process.svg](resource/migration_process.svg)
+![migration_process.svg](https://raw.gitcode.com/SMAT/HomeTrans/raw/main/resource/migration_process.svg)
 
 ---
 
@@ -132,7 +97,7 @@
 
 ### 1.1UI迁移(`hmos-batch-ui-align`)
 
-> 📦 依赖:python(必需,页面快照解析脚本);Android Studio + 安卓真机/模拟器(可选,仅自动抓取页面快照时需要)。
+> 📦 依赖:uv(必需,页面快照解析脚本经 `uv run` 运行,uv 自带 Python);Android Studio + 安卓真机/模拟器(可选,仅自动抓取页面快照时需要)。
 
 ```
 /hmos-batch-ui-align android_project_dir=<安卓项目路径> harmony_project_dir=<鸿蒙项目路径> ui_info_root=<页面快照目录>
@@ -157,7 +122,7 @@
 
 ### 1.2UI对齐(`hmos-incremental-ui-align`,按需)
 
-> 📦 依赖:DevEco Studio、Android Studio、python(必需,双端页面采集脚本);安卓真机/模拟器 + 鸿蒙真机/模拟器(必需)。
+> 📦 依赖:DevEco Studio、Android Studio、uv(必需,双端页面采集脚本经 `uv run` 运行,uv 自带 Python);安卓真机/模拟器 + 鸿蒙真机/模拟器(必需)。
 
 ```
 /hmos-incremental-ui-align android_project_dir=<安卓工程路径> harmony_project_dir=<鸿蒙工程路径>
@@ -183,7 +148,7 @@
 
 ## 步骤2:生成需求规格(`hmos-spec-generate`)
 
-> 📦 依赖:GitNexus(必需)。
+> 📦 依赖:homegraph(必需)。
 
 > 👤 前置:用户编写 `REQ.txt` 文件 — 以自然语言描述原始需求，如有多个需求则在文件中使用空行进行分隔，需求中建议包含页面跳转路径的说明以及业务逻辑的描述。
 
@@ -231,13 +196,13 @@
 | 产物 | 位置 | 说明 |
 |------|------|------|
 | `<feature>-SPEC.md` | `spec_output_dir` | 每个REQ对应一份原子场景规格文档 |
-| `.trace/<feature>.md` | `spec_output_dir/.trace/` | 每个REQ对应的代码trace(GitNexus探索结果) |
+| `.trace/<feature>.md` | `spec_output_dir/.trace/` | 每个REQ对应的代码trace(homegraph探索结果) |
 
 ---
 
 ## 步骤3:逻辑代码转换流水线(`hmos-convert-pipeline`)
 
-> 📦 依赖:DevEco Studio(必需,构建/评审修复);uv + 鸿蒙真机(集成测试阶段需要,可经 `skip_test` 跳过)。
+> 📦 依赖:DevEco Studio(必需,构建/评审修复);uv + 鸿蒙真机(集成测试阶段需要,可经 `skip_test` 跳过;uv 经 `uv run` 运行 `self_test_runner.py` 并自带 Python)。
 
 > 👤 前置:
 > 1. 如果执行集成测试，需要**编写测试用例文档**输出前置测试用例（非必须，用于测试用例的环境准备等）通过pre_test_case_path参数传递,测试用例通过test_case_path参数传递。

package/agents/build-fixer.md

@@ -29,24 +29,25 @@ Execute a build-fix loop: resolve environment → validate project → build →
 
 The build requires the **DevEco Studio installation path**, from which Node.js, hvigor, and ohpm are derived.
 
-### 0a. Read the DevEco environment variables
+### 0a. Read the DevEco config values
 
-Read these **directly from the OS environment** — `ht init` persists them as machine environment variables (`echo "$VAR"` on macOS/Linux; `$env:VAR` in PowerShell on Windows):
+Resolve each value along the standard chain — **OS environment variable first, then `~/.hometrans/config.json` (its `env.*` block), then ask the user**. `ht init` persists these both as machine environment variables and in config.json (read env vars with `echo "$VAR"` on macOS/Linux; `$env:VAR` in PowerShell on Windows):
 
-| Variable | Meaning |
-|---|---|
-| `DEVECO_SDK_HOME` | DevEco sdk dir, e.g. `D:/DevEco Studio/sdk` — single source of truth |
-| `DEVECO_HOME` | DevEco install root, e.g. `D:/DevEco Studio` |
-| `OHOS_SDK_PATH` | `<DEVECO_SDK_HOME>/default/openharmony/ets` |
-| `HMS_SDK_PATH` | `<DEVECO_SDK_HOME>/default/hms/ets` |
+| Variable | config.json fallback | Meaning |
+|---|---|---|
+| `DEVECO_SDK_HOME` | `env.DEVECO_SDK_HOME` | DevEco sdk dir, e.g. `D:/DevEco Studio/sdk` — single source of truth |
+| `DEVECO_HOME` | `env.DEVECO_HOME` | DevEco install root, e.g. `D:/DevEco Studio` |
+| `OHOS_SDK_PATH` | `env.OHOS_SDK_PATH` | `<DEVECO_SDK_HOME>/default/openharmony/ets` |
+| `HMS_SDK_PATH` | `env.HMS_SDK_PATH` | `<DEVECO_SDK_HOME>/default/hms/ets` |
 
-### 0b. Resolve the DevEco root `<deveco>` — env vars first, then auto-detect, then ask
+### 0b. Resolve the DevEco root `<deveco>` — env vars → config.json → auto-detect → ask
 
 Stop at the **first** source that yields a path:
 
-1. **Environment variables**: `DEVECO_HOME` if non-empty → that is `<deveco>`; else `DEVECO_SDK_HOME` → `<deveco>` is its **parent directory**; else (legacy) strip the trailing `/sdk/default/openharmony/ets` from `OHOS_SDK_PATH` (or `/sdk/default/hms/ets` from `HMS_SDK_PATH`) to get `<deveco>`.
-2. **Auto-detect** (0c below).
-3. If both fail, **ask the user** for the DevEco Studio install path (suggest running `ht init` to persist it as an environment variable).
+1. **OS environment variables**: `DEVECO_HOME` if non-empty → that is `<deveco>`; else `DEVECO_SDK_HOME` → `<deveco>` is its **parent directory**; else (legacy) strip the trailing `/sdk/default/openharmony/ets` from `OHOS_SDK_PATH` (or `/sdk/default/hms/ets` from `HMS_SDK_PATH`) to get `<deveco>`.
+2. **`~/.hometrans/config.json`**: apply the same derivation to `env.DEVECO_HOME` / `env.DEVECO_SDK_HOME` (read with `uv run --no-project python -c "import json;from pathlib import Path;p=Path.home()/'.hometrans'/'config.json';print((json.loads(p.read_text(encoding='utf-8')).get('env') or {}).get('DEVECO_HOME','') if p.exists() else '')"`).
+3. **Auto-detect** (0c below).
+4. If all fail, **ask the user** for the DevEco Studio install path (suggest running `ht init` to persist it as an environment variable).
 
 > Example: `DEVECO_SDK_HOME = D:/DevEco Studio/sdk` → `<deveco> = D:/DevEco Studio`.
 
@@ -58,9 +59,9 @@ From `<deveco>`, derive:
 
 Verify these files exist. If they do, use them and skip auto-detection. If an env-sourced path does **not** exist on disk, treat that source as invalid and continue down the resolution order.
 
-### 0c. Auto-detect if env vars are missing or invalid
+### 0c. Auto-detect if env vars and config.json are missing or invalid
 
-If the environment variables do not yield a valid install, auto-detect:
+If neither the environment variables nor `~/.hometrans/config.json` yield a valid install, auto-detect:
 
 1. **Find DevEco Studio installation**:
    - Search common locations: `D:\DevEco Studio`, `C:\DevEco Studio`, `C:\Program Files\DevEco Studio`

package/agents/code-reviewer.md

@@ -33,10 +33,10 @@ Before starting the review, extract the code context for the commit via the `ext
    - `projectPath`: `<harmony_project_dir>` (absolute path to the HarmonyOS project root containing the `.git` directory)
    - `commitId`: `<commit_id>` (the git commit to analyze — diffs against its first parent)
    - `mode`: `"default"` (builds full call graph for call-chain context)
-   - `ohosSdkPath` / `hmsSdkPath`: the OpenHarmony / HMS SDK ETS directory paths. Resolve each **from the OS environment**, stopping at the first hit:
+   - `ohosSdkPath` / `hmsSdkPath`: the OpenHarmony / HMS SDK ETS directory paths. Resolve each along the standard chain — **OS environment variable → `~/.hometrans/config.json` → ask the user** — stopping at the first hit:
      1. `OHOS_SDK_PATH` / `HMS_SDK_PATH` environment variables; if those are empty but `DEVECO_SDK_HOME` is set, derive them as `<DEVECO_SDK_HOME>/default/openharmony/ets` and `<DEVECO_SDK_HOME>/default/hms/ets`.
-     2. Omit the parameter to let the tool read `OHOS_SDK_PATH` / `HMS_SDK_PATH` from the environment itself.
-     3. If the environment provides no path that exists on disk, **ask the user** for the DevEco `sdk` directory (and suggest running `ht init` to persist it as an environment variable).
+     2. Omit the parameter to let the `extract_commit_context` tool resolve them itself — it reads `OHOS_SDK_PATH` / `HMS_SDK_PATH` from the OS environment, then falls back to `env.OHOS_SDK_PATH` / `env.HMS_SDK_PATH` in `~/.hometrans/config.json`.
+     3. If neither the environment nor `~/.hometrans/config.json` provides a path that exists on disk, **ask the user** for the DevEco `sdk` directory (and suggest running `ht init` to persist it as an environment variable).
 
 2. **On success** — the tool returns the affected code context (diffs, call graphs, impacted files). Treat this response as the **code context** for the review.
 

package/agents/logic-coder.md

@@ -16,7 +16,7 @@ min-model: sonnet
 
 Platform context query:
 
-    python3 {scripts}/platform_context_query.py \
+    uv run --no-project python {scripts}/platform_context_query.py \
       --request {output_path}/platform-context-request.json \
       --out-dir {output_path}/platform-context-{N}
 

package/agents/logic-context-builder.md

@@ -16,7 +16,7 @@ min-model: sonnet
 
 Platform context query:
 
-    python3 {scripts}/platform_context_query.py \
+    uv run --no-project python {scripts}/platform_context_query.py \
       --request {output_path}/platform-context-request.json \
       --out-dir {output_path}/platform-context-{N}
 

package/agents/self-tester.md

@@ -38,16 +38,19 @@ You are a self-tester for HarmonyOS applications. You run a single pipeline: par
 
 ### AutoTest Directory Locator
 
-Used independently by S5 and T4. The AutoTest tools are installed by `ht init` under the HomeTrans tools dir; resolve their location from config rather than searching the filesystem.
+Used independently by S5 and T4. The AutoTest tools are installed by `ht init` under the HomeTrans tools dir; resolve their location along the standard config chain — **OS environment variable → `~/.hometrans/config.json` → filesystem walk-up** — rather than searching the filesystem first.
 
-**Primary — read `HOMETRANS_TOOL_PATH` directly from the OS environment.** The autotest dir is `<HOMETRANS_TOOL_PATH>/test-tools/autotest`:
+**Tier 1 — OS environment variable `HOMETRANS_TOOL_PATH`; Tier 2 — `env.HOMETRANS_TOOL_PATH` in `~/.hometrans/config.json`.** The autotest dir is `<tool_path>/test-tools/autotest`:
 
 ```bash
 tool_path="$HOMETRANS_TOOL_PATH"
+if [ -z "$tool_path" ]; then \
+  tool_path="$(uv run --no-project python -c "import json,os; from pathlib import Path; cfg=Path.home()/'.hometrans'/'config.json'; print((json.loads(cfg.read_text(encoding='utf-8')).get('env') or {}).get('HOMETRANS_TOOL_PATH','') if cfg.exists() else '')" 2>/dev/null)"; \
+fi
 if [ -n "$tool_path" ] && [ -d "$tool_path/test-tools/autotest" ]; then echo "$tool_path/test-tools/autotest"; fi
 ```
 
-If the `HOMETRANS_TOOL_PATH` environment variable is unset or invalid, **fall back** to walking up from `<output_path>` for a sibling `test-tools/autotest` (then `agents/test-tools/autotest` for legacy layouts) — **anchor the search at `<output_path>`, the sub-agent cwd is not guaranteed**:
+If neither the env var nor `~/.hometrans/config.json` yields a valid tool path, **fall back** (Tier 3) to walking up from `<output_path>` for a sibling `test-tools/autotest` (then `agents/test-tools/autotest` for legacy layouts) — **anchor the search at `<output_path>`, the sub-agent cwd is not guaranteed**:
 
 ```bash
 d="<output_path>"; while [ "$d" != "/" ] && [ "$d" != "" ]; do \
@@ -195,7 +198,7 @@ cd "<autotest_dir>" && uv run python testcases_tool.py generate "<output_path>/_
 **Sanity-check `[PRE]` ordering** — run this command **if and only if** a `pre_test_case.md` was found and extracted in S4.1. If no pre-cases were extracted, skip this step entirely (the assertion would spuriously fail on an empty `pre` list):
 
 ```bash
-python -c "import json,sys; d=json.load(open(sys.argv[1],encoding='utf-8')); pre=[i for i,c in enumerate(d) if c['case_name'].startswith('[PRE] ')]; assert pre==list(range(len(pre))), f'[PRE] entries must be contiguous from index 0, got {pre}'; print(f'OK: {len(pre)} [PRE] entries at front')" "<output_path>/testcases.json"
+uv run --no-project python -c "import json,sys; d=json.load(open(sys.argv[1],encoding='utf-8')); pre=[i for i,c in enumerate(d) if c['case_name'].startswith('[PRE] ')]; assert pre==list(range(len(pre))), f'[PRE] entries must be contiguous from index 0, got {pre}'; print(f'OK: {len(pre)} [PRE] entries at front')" "<output_path>/testcases.json"
 ```
 
 If assertion fails, edit `_extracted.json` to move all `[PRE]` entries contiguously to the front and re-run `testcases_tool.py`.
@@ -219,7 +222,7 @@ test -f "<hap_path>" && mkdir -p "<output_path>" && test -f "<output_path>/testc
 If `setup=false`, additionally verify each JSON parses and `testcases.json` is non-empty:
 
 ```bash
-python -c "import json,sys; t=json.load(open(sys.argv[1],encoding='utf-8')); m=json.load(open(sys.argv[2],encoding='utf-8')); assert isinstance(t,list) and len(t)>0, 'testcases.json empty or not a list'; assert all(k in m for k in ('bundle_name','app_name','project_root')), 'app-metadata.json missing required fields'; print('OK')" "<output_path>/testcases.json" "<output_path>/app-metadata.json"
+uv run --no-project python -c "import json,sys; t=json.load(open(sys.argv[1],encoding='utf-8')); m=json.load(open(sys.argv[2],encoding='utf-8')); assert isinstance(t,list) and len(t)>0, 'testcases.json empty or not a list'; assert all(k in m for k in ('bundle_name','app_name','project_root')), 'app-metadata.json missing required fields'; print('OK')" "<output_path>/testcases.json" "<output_path>/app-metadata.json"
 ```
 
 If any check fails, write a `self-test-report.md` using the sentinel format from Shared Utilities. The `reason:` line names the specific failure, e.g.:
@@ -246,17 +249,17 @@ Record the device serial number.
 
 ### T4 — Locate AutoTest Directory & Verify autotest config
 
-Find the `test-tools/autotest` directory using the **AutoTest Directory Locator** in Shared Utilities (OS env var `HOMETRANS_TOOL_PATH` → walk up from `<output_path>`). If not found, write a `self-test-report.md` with the sentinel format (`status: FAIL` / `reason: AutoTest directory not found from HOMETRANS_TOOL_PATH env var or output_path`), then stop.
+Find the `test-tools/autotest` directory using the **AutoTest Directory Locator** in Shared Utilities (OS env var `HOMETRANS_TOOL_PATH` → `env.HOMETRANS_TOOL_PATH` in `~/.hometrans/config.json` → walk up from `<output_path>`). If not found, write a `self-test-report.md` with the sentinel format (`status: FAIL` / `reason: AutoTest directory not found from HOMETRANS_TOOL_PATH env var or output_path`), then stop.
 
 Set `<autotest_dir>` to the found path.
 
-The AutoTest configuration now lives in the `autotest` block of `~/.hometrans/config.json` (it replaced the old `config.yaml`). `self_test_runner.py` reads it at run time and materializes a transient YAML for `AutoTest.batch` — you do **not** create or edit any YAML. Just verify the config is usable: the `autotest` block exists with a real `api_key`, **or** the `TEST_API_KEY` OS environment variable is set (the runner uses it to fill a placeholder/empty key).
+The model config is shared with UI alignment and resolved along the standard chain — **env var → `~/.hometrans/config.json` → ask**. `self_test_runner.py` reads it at run time (env `HOMETRANS_MODEL_API_KEY` overrides `autotest.unified_model.api_key`; legacy `TEST_API_KEY` still honored as a fallback) and materializes a transient YAML for `AutoTest.batch` — you do **not** create or edit any YAML. Just verify the api_key is usable: the `HOMETRANS_MODEL_API_KEY` (or legacy `TEST_API_KEY`) OS environment variable is set, **or** the `autotest` block in `~/.hometrans/config.json` has a real (non-placeholder) `api_key`.
 
 ```bash
-python -c "import json,os,sys; from pathlib import Path; cfg=Path.home()/'.hometrans'/'config.json'; tp=os.environ.get('HOMETRANS_TOOL_PATH'); cand=(Path(tp).parent/'config.json') if tp else None; cfg=cand if (cand and cand.exists()) else cfg; at=(json.loads(cfg.read_text(encoding='utf-8')).get('autotest') or {}) if cfg.exists() else {}; real=any(isinstance(at.get(k),dict) and str(at[k].get('api_key','')).strip() not in ('','YOUR_API_KEY_HERE') for k in ('unified_model','execute_model','decision_model')); print('OK' if (real or os.environ.get('TEST_API_KEY','').strip()) else 'MISSING')"
+uv run --no-project python -c "import json,os,sys; from pathlib import Path; cfg=Path.home()/'.hometrans'/'config.json'; tp=os.environ.get('HOMETRANS_TOOL_PATH'); cand=(Path(tp).parent/'config.json') if tp else None; cfg=cand if (cand and cand.exists()) else cfg; at=(json.loads(cfg.read_text(encoding='utf-8')).get('autotest') or {}) if cfg.exists() else {}; real=any(isinstance(at.get(k),dict) and str(at[k].get('api_key','')).strip() not in ('','YOUR_API_KEY_HERE') for k in ('unified_model','execute_model','decision_model')); env=os.environ.get('HOMETRANS_MODEL_API_KEY','').strip() or os.environ.get('TEST_API_KEY','').strip(); print('OK' if (env or real) else 'MISSING')"
 ```
 
-If this does not print `OK` (no `autotest` block with a real `api_key`, and no `TEST_API_KEY` in the OS environment), write `self-test-report.md` with the sentinel format (`status: FAIL` / `reason: autotest api_key not configured — set TEST_API_KEY env var or run ht init`), then stop.
+If this does not print `OK` (no `HOMETRANS_MODEL_API_KEY` in the OS environment, and no `autotest` block with a real `api_key`), write `self-test-report.md` with the sentinel format (`status: FAIL` / `reason: model api_key not configured — set HOMETRANS_MODEL_API_KEY env var or run ht init`), then stop.
 
 ### T5 — Clean Task Directory
 
@@ -267,7 +270,7 @@ rm -rf "<output_path>/task" && echo "Task directory cleaned" || echo "Task direc
 ### T6 — Run self_test_runner.py
 
 > 🚨 **MANDATORY**: Environment setup, HAP installation, and test execution are all handled by a **single invocation** of `self_test_runner.py run`. The script synchronously executes:
-> 1. Read & validate the `autotest` config from `~/.hometrans/config.json` (filling the api_key from `TEST_API_KEY` if needed) and materialize a transient YAML for AutoTest.batch
+> 1. Read & validate the `autotest` config from `~/.hometrans/config.json` (env `HOMETRANS_MODEL_API_KEY` overrides the api_key; legacy `TEST_API_KEY` honored as a fallback) and materialize a transient YAML for AutoTest.batch
 > 2. `uv sync` — installs `harmony-autotest` + `hypium_mcp`
 > 3. `hdc uninstall` + `hdc install -r`
 >
@@ -281,7 +284,7 @@ rm -rf "<output_path>/task" && echo "Task directory cleaned" || echo "Task direc
 > - ❌ Writing a shell loop or Python loop to iterate over cases yourself
 
 ```bash
-cd "<autotest_dir>" && python self_test_runner.py run --testcases "<output_path>/testcases.json" --hap "<hap_path>" --bundle-name "<bundle_name>" --category "<app_name>" --task-dir "<output_path>/task" --output-dir "<output_path>"
+cd "<autotest_dir>" && uv run --no-project python self_test_runner.py run --testcases "<output_path>/testcases.json" --hap "<hap_path>" --bundle-name "<bundle_name>" --category "<app_name>" --task-dir "<output_path>/task" --output-dir "<output_path>"
 ```
 
 Record the `pid` and `task_subdir` from the output.
@@ -291,7 +294,7 @@ Record the `pid` and `task_subdir` from the output.
 Each poll MUST include `sleep 60`. **Use `timeout: 120000` on each Bash call** so the 60-second sleep plus the status command have headroom and the status check itself isn't truncated:
 
 ```bash
-sleep 60 && cd "<autotest_dir>" && python self_test_runner.py status --task-dir "<output_path>/task" --output-dir "<output_path>"
+sleep 60 && cd "<autotest_dir>" && uv run --no-project python self_test_runner.py status --task-dir "<output_path>/task" --output-dir "<output_path>"
 ```
 
 | `status` | Meaning | Exit code | Action |
@@ -310,7 +313,7 @@ sleep 60 && cd "<autotest_dir>" && python self_test_runner.py status --task-dir
 
 **Polling rules:**
 - Max iterations: **N_cases × 6** (each case takes ~6 minutes). For 8 cases → max 48 polls. If still `RUNNING` after that, kill and write TIMEOUT report.
-- **Timeout kill**: `cd "<autotest_dir>" && python self_test_runner.py kill --task-dir "<output_path>/task"`. This verifies the PID belongs to an `AutoTest.batch` process before terminating it.
+- **Timeout kill**: `cd "<autotest_dir>" && uv run --no-project python self_test_runner.py kill --task-dir "<output_path>/task"`. This verifies the PID belongs to an `AutoTest.batch` process before terminating it.
 - **CRASHED**: Write CRASHED report with sentinel (`status: FAIL` / `reason: AutoTest batch crashed`). Include the `log_tail` from the status response. `task_results.jsonl` is unavailable.
 - **TIMEOUT**: After killing, write TIMEOUT report with sentinel (`status: FAIL` / `reason: AutoTest batch timed out after N polls`) and include partial results from `task_results.jsonl` if any cases finished.