@buaa_smat/hometrans 0.1.7 → 0.1.9

package/package.json

@@ -1,8 +1,12 @@
 {
   "name": "@buaa_smat/hometrans",
-  "version": "0.1.7",
+  "version": "0.1.9",
   "description": "HomeTrans (Android-to-HarmonyOS) skill + agent installer. Run `ht init` to distribute conversion skills and subagents into AI editors.",
   "license": "MIT",
+  "repository": {
+    "type": "git",
+    "url": "https://gitcode.com/SMAT/HomeTrans"
+  },
   "type": "module",
   "bin": {
     "hometrans": "dist/cli/index.js",
@@ -12,7 +16,8 @@
     "dist",
     "skills",
     "agents",
-    "tools"
+    "tools",
+    "resource"
   ],
   "scripts": {
     "build": "node scripts/build.js",

package/skills/hmos-convert-pipeline/SKILL.md

@@ -20,10 +20,9 @@ Parse `$ARGUMENTS` as positional tokens:
 - **Arg 3** (`assets-output-path`): Directory to store all output/report files (required)
 - **Arg 4** (`max-rounds-review`): Maximum number of Stage 3→3a→3b code-review-fix rounds to run (optional, default `2`). Must be a positive integer `>= 1`.
 - **Arg 5** (`max-rounds-test`): Maximum number of Stage 4→4a→4b self-test rounds to run (optional, default `2`). Must be a positive integer `>= 1`.
-- **Arg 6** (`variant`): `enhanced` or `baseline` (optional, default `enhanced`). Selects the Stage 1 / 1a agent family. `enhanced` uses the full `logic-context-builder` + `logic-coder` agents; `baseline` uses the pure-LLM `logic-context-builder-minimal` + `logic-coding-minimal` agents for A/B capability comparison. All other stages and their agents are unchanged. To pass this arg, provide explicit values for Args 4 and 5 first.
-- **Arg 7** (`skip-test`): `true` or `false` (optional, default `false`). When `true`, skip Stage 4 / 4a / 4b (Self-Testing Loop) entirely. Use this when no real HarmonyOS device is available for on-device testing. To pass this arg, provide explicit values for Args 4, 5, and 6 first.
+- **Arg 6** (`skip-test`): `true` or `false` (optional, default `false`). When `true`, skip Stage 4 / 4a / 4b (Self-Testing Loop) entirely. Use this when no real HarmonyOS device is available for on-device testing. To pass this arg, provide explicit values for Args 4 and 5 first.
 
-If any required argument is missing, ask the user before proceeding. If `max-rounds-review` or `max-rounds-test` is provided but is not a positive integer, ask the user before proceeding. If `variant` is provided but is not `enhanced` or `baseline`, ask the user before proceeding. If `skip-test` is provided but is not `true` or `false`, ask the user before proceeding.
+If any required argument is missing, ask the user before proceeding. If `max-rounds-review` or `max-rounds-test` is provided but is not a positive integer, ask the user before proceeding. If `skip-test` is provided but is not `true` or `false`, ask the user before proceeding.
 
 Define shorthand variables for the instructions below:
 
@@ -34,7 +33,6 @@ Define shorthand variables for the instructions below:
 | `OUTPUT` | assets-output-path (resolved or generated) |
 | `MAX_ROUNDS_REVIEW` | max-rounds-review — positive integer (default `2`). Controls Stage 3→3a→3b code-review-fix loop. |
 | `MAX_ROUNDS_TEST`   | max-rounds-test — positive integer (default `2`). Controls Stage 4→4a→4b self-test loop. |
-| `VARIANT` | variant — `enhanced` (default) or `baseline`. `baseline` switches Stage 1/1a to the minimal pure-LLM baseline agents. |
 | `SKIP_TEST` | skip-test — `true` or `false` (default `false`). When `true`, skip Stage 4 / 4a / 4b entirely (no real device available). |
 
 ---
@@ -145,40 +143,26 @@ Flags: `skip-plan-builder: true|false` (default `false`) — when `true`, skip t
 
 Prompt format (applies to both Stage 1 and Stage 1a): ONLY the key-value lines below. No natural language, step lists, or Markdown.
 
-1. Launch the logic context builder agent. Pick by `VARIANT`:
+1. Launch the logic context builder agent:
    ```
-   # VARIANT=enhanced (default) — full agent
    Agent(
      subagent_type="logic-context-builder",
      prompt="harmonyos-project-path: HMOS\nspec-file: OUTPUT/plan.md\noutput-path: OUTPUT/logic"
    )
-
-   # VARIANT=baseline — pure-LLM baseline agent
-   Agent(
-     subagent_type="logic-context-builder-minimal",
-     prompt="harmonyos-project-path: HMOS\nspec-file: OUTPUT/plan.md\noutput-path: OUTPUT/logic"
-   )
    ```
 2. Verify `OUTPUT/logic/plan.md` exists.
 
 ### Stage 1a — Logic Coding
 
-1. Launch the logic coding agent. Pick by `VARIANT`:
+1. Launch the logic coding agent:
    ```
-   # VARIANT=enhanced (default) — full agent
    Agent(
      subagent_type="logic-coder",
      prompt="harmonyos-project-path: HMOS\nplan-file: OUTPUT/logic/plan.md\noutput-path: OUTPUT/logic"
    )
-
-   # VARIANT=baseline — pure-LLM baseline agent
-   Agent(
-     subagent_type="logic-coding-minimal",
-     prompt="harmonyos-project-path: HMOS\nplan-file: OUTPUT/logic/plan.md\noutput-path: OUTPUT/logic"
-   )
    ```
 2. Copy `OUTPUT/logic/commit-info.md` → `OUTPUT/commit-info.md` and verify it contains a hex `commit-id`.
-3. `OUTPUT/logic/coding-summary.md` (enhanced mode only) is human-readable auxiliary; downstream stages consume only `OUTPUT/commit-info.md`.
+3. `OUTPUT/logic/coding-summary.md` is human-readable auxiliary; downstream stages consume only `OUTPUT/commit-info.md`.
 
 ### Stage 2 — Compilation and Build
 

package/skills/hmos-integration-test/SKILL.md

@@ -30,11 +30,11 @@ Extract the following from the user's message. If any required input is missing,
 | Variable | Required | Meaning | Typical Phrasing |
 |----------|----------|---------|------------------|
 | `test-case-path` | yes | Path to test_case.md | "测试用例在...", "用例文件...", "test case path" |
-| `output-path` | no (default: directory of `test-case-path`) | Directory for all output artifacts | "输出到...", "产物目录...", "output to..." |
 | `hap-path` | yes | Path to the signed .hap file | "HAP路径...", "安装包在...", "hap file..." |
-| `pre-test-case-path` | optional | Path to pre_test_case.md | "前置用例...", "pre test case..." |
-| `android-project-path` | optional (for fix loop) | Path to Android source project for reference-based fixing | "Android路径...", "android project..." |
-| `max-rounds` | optional, default `3` | Max iterations for the test-and-fix loop. Must be a positive integer (`>= 1`). Only meaningful when the fix loop is enabled. | "最多 X 轮", "max rounds X", "迭代 X 次" |
+| `output-path` | no (default: directory of `test-case-path`) | Directory for all output artifacts | "输出到...", "产物目录...", "output to..." |
+| `pre-test-case-path` | no | Path to pre_test_case.md | "前置用例...", "pre test case..." |
+| `android-project-path` | no (for fix loop) | Path to Android source project for reference-based fixing | "Android路径...", "android project..." |
+| `max-rounds` | no (default `3`) | Max iterations for the test-and-fix loop. Must be a positive integer (`>= 1`). Only meaningful when the fix loop is enabled. | "最多 X 轮", "max rounds X", "迭代 X 次" |
 
 **AskUserQuestion template (use this exact shape — do NOT ad-lib free-form questions):**
 

package/tools/test-tools/autotest/README.md

@@ -18,7 +18,7 @@ HarmonyOS 应用自测执行器。负责把自然语言写的测试用例（`tes
 | `hdc` | 设备连接 / 安装 HAP | `hdc version` |
 | `python` ≥ 3.10 | 入口脚本 | `python --version` |
 
-> `uv sync`（拉取 6 个 whl）由 `self_test_runner.py` 自动执行，无需手动 `pip install`。
+> `uv sync`（从本地 `tools/test-tools/` 目录加载 6 个 whl）由 `self_test_runner.py` 自动执行，无需手动 `pip install`。
 
 ### 2. 填写 `config.yaml`（包含真实 api_key）
 
@@ -183,7 +183,7 @@ uv run python report_tool.py validate <out>/self-test-report.md
 agents/test-tools/autotest/
 ├── README.md                       # 本文件
 ├── .gitignore                      # 排除 .venv / config.yaml / 运行期日志等
-├── pyproject.toml                  # 声明 6 个远程 whl URL（harmony-autotest / hypium_mcp / hypium / xdevice*3）
+├── pyproject.toml                  # 声明 6 个本地 whl 路径（harmony-autotest / hypium_mcp / hypium / xdevice*3，均在上一级 tools/test-tools/）
 ├── uv.lock                         # 与 pyproject.toml 配套的版本锁定文件
 ├── config.yaml.example             # 用户拷贝填 api_key 的模板
 ├── config.yaml                     # 用户填好的真实配置（.gitignore 中，不入库）
@@ -202,13 +202,13 @@ agents/test-tools/autotest/
 A: 复制 `config.yaml.example` → `config.yaml` 并填入真实 api_key 后重试。
 
 **Q: `uv sync` 卡住或失败**
-A: 6 个 whl 都从 gitcode API v5 拉，需要外网访问 `gitcode.com`。失败时检查代理 / 网络；也可以 `--force-reinit` 强制重建。
+A: 6 个 whl 改为从本地 `tools/test-tools/` 目录加载（`pyproject.toml` 的 `[tool.uv.sources]` 用相对路径 `../*.whl`）。失败时确认这 6 个 whl 文件存在于上一级目录；其余依赖仍从清华 PyPI 镜像拉取，需基本网络。也可以 `--force-reinit` 强制重建。
 
 **Q: 测试过程被打断，下次跑会冲突吗？**
 A: 不会。`self_test_runner.py run` 启动时会扫描并 kill 残留的 `AutoTest.batch` / `hypium_mcp` / `harmony_autotest` 进程（不止靠 PID 文件，按命令行名兜底），并自动清掉本目录顶层的 `tmp_hypium/` / `hypium_mcp.log` 等遗留中间产物。`task/` 目录在 agent 流程里也会被清空。
 
 **Q: 想彻底重置一遍本地状态再跑？**
-A: 跑一次 `python self_test_runner.py clean`：杀残留 → 删 `.venv` → 删 `tmp_hypium/` / `hypium_mcp.log` / `batch.pid` / `batch_stdout.log`。**不要**手工 `uv cache clean`——全局缓存清空后重装 6 个 whl 要从 gitcode 重新拉（1–3 分钟）；保留缓存的话 `uv sync` 只需几秒。
+A: 跑一次 `python self_test_runner.py clean`：杀残留 → 删 `.venv` → 删 `tmp_hypium/` / `hypium_mcp.log` / `batch.pid` / `batch_stdout.log`。**不要**手工 `uv cache clean`——全局缓存清空后重装 6 个 whl 要从 gitcode 重新拉（1–3 分钟）；这 6 个 whl 现在从本地加载、不走网络，但其余依赖仍走缓存，保留缓存的话 `uv sync` 只需几秒。
 
 **Q: 想自己改 `testcases.json` 后再跑？**
 A: 改完后跑一遍校验，再交给 `self-tester`（`setup=false`）：

package/tools/test-tools/autotest/pyproject.toml

@@ -4,13 +4,23 @@ version = "0.1.0"
 description = "Self-test runner harness for HarmonyOS apps; delegates execution to the harmony-autotest whl."
 requires-python = ">=3.10"
 dependencies = [
-    "harmony-autotest @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/harmony_autotest-0.1.0-py3-none-any.whl?ref=main",
-    "hypium_mcp @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/hypium_mcp-0.6.5-py3-none-any.whl?ref=main",
-    "hypium @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/hypium-6.1.0.210-py3-none-any.whl?ref=main",
-    "xdevice @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/xdevice-6.1.0.210-py3-none-any.whl?ref=main",
-    "xdevice-ohos @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/xdevice_ohos-6.1.0.210-py3-none-any.whl?ref=main",
-    "xdevice-devicetest @ https://gitcode.com/api/v5/repos/SMAT/HomeTrans/raw/agents/test-tools/xdevice_devicetest-6.1.0.210-py3-none-any.whl?ref=main",
+    "harmony-autotest",
+    "hypium_mcp",
+    "hypium",
+    "xdevice",
+    "xdevice-ohos",
+    "xdevice-devicetest",
 ]
 
+# 6 个 whl 改为从本地加载（仓库里放在上一级 tools/test-tools/ 目录），
+# 路径相对于本 pyproject.toml；不再从 gitcode 远程拉取。
+[tool.uv.sources]
+harmony-autotest = { path = "../harmony_autotest-0.1.0-py3-none-any.whl" }
+hypium_mcp = { path = "../hypium_mcp-0.6.5-py3-none-any.whl" }
+hypium = { path = "../hypium-6.1.0.210-py3-none-any.whl" }
+xdevice = { path = "../xdevice-6.1.0.210-py3-none-any.whl" }
+xdevice-ohos = { path = "../xdevice_ohos-6.1.0.210-py3-none-any.whl" }
+xdevice-devicetest = { path = "../xdevice_devicetest-6.1.0.210-py3-none-any.whl" }
+
 [[tool.uv.index]]
 url = "https://pypi.tuna.tsinghua.edu.cn/simple"

package/tools/test-tools/autotest/self_test_runner.py

@@ -22,8 +22,8 @@
             ↓
     后台：python -m AutoTest.batch（循环执行 testcases.jsonl，cwd 为 task_<timestamp> 子目录）
 
-依赖：harmony-autotest 与 hypium_mcp 均在 pyproject.toml 中声明为远程 whl URL
-依赖；`uv sync` 会拉取到本目录 .venv 内，不再需要手动 pip install。
+依赖：harmony-autotest 与 hypium_mcp 等 6 个 whl 在 pyproject.toml 中声明为本地路径
+依赖（上一级 tools/test-tools/ 目录）；`uv sync` 会安装到本目录 .venv 内，不再需要手动 pip install。
 
 AutoTest.batch 启动时 cwd 设为 task_<timestamp>，因此 hypium_mcp.log / tmp_hypium/
 等中间产物会落在 task 子目录内，避免污染插件源码目录。
@@ -707,8 +707,8 @@ def cmd_clean(args):
       - 删除 .venv 及 __pycache__
       - 删除本目录顶层的 tmp_hypium / hypium_mcp.log / batch.pid / batch_stdout.log
 
-    NOT touched: 全局 uv 缓存。重建缓存需重新从 gitcode 拉 6 个 whl，
-    单纯"重跑一遍"不需要碰它。需要的话用 `uv cache clean` 自己清。
+    NOT touched: 全局 uv 缓存。6 个 whl 已改为从本地加载，不走网络；
+    单纯"重跑一遍"不需要碰缓存。需要的话用 `uv cache clean` 自己清。
     """
     # 控制台输出即可，不写 self_test_*.log（避免污染 output-dir）
     logging.basicConfig(