openmatrix 0.1.8 → 0.1.10

package/README.md

@@ -358,6 +358,23 @@ Accept 阶段由 Reviewer Agent 执行:
 
 ---
 
+## 多语言支持
+
+OpenMatrix 通过 Claude Code Agent 工具**原生支持所有主流编程语言**：
+
+| 语言 | 测试命令 | 构建命令 |
+|------|---------|---------|
+| TypeScript/JavaScript | `npm test` / `vitest` | `npm run build` |
+| Python | `pytest` | `python -m build` |
+| Go | `go test ./...` | `go build` |
+| Java | `mvn test` | `mvn compile` |
+| Rust | `cargo test` | `cargo build` |
+| 其他 | 任意 CLI 命令 | 任意 CLI 命令 |
+
+**无需额外配置** — Agent 可执行任意 shell 命令，Claude 理解所有主流语言。
+
+---
+
 ## 常见问题
 
 ### Q: 哪种质量级别适合我？
@@ -414,9 +431,8 @@ cd openmatrix && npm install && npm run build && npm test
 - [x] 质量报告
 - [x] AI 验收
 - [x] `/om:auto` 全自动模式
-- [x] Meeting 自动处理
+- [x] 多语言支持 (Python/Go/Java/TypeScript 等)
 - [ ] VSCode 扩展
-- [ ] Python/Go 支持
 - [ ] CI/CD 集成
 
 ---

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "openmatrix",
-  "version": "0.1.8",
+  "version": "0.1.10",
   "description": "AI Agent task orchestration system with Claude Code Skills integration",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/skills/auto.md

@@ -7,6 +7,21 @@ description: 全自动执行任务 - 无阻塞，bypass permissions
 执行此技能时，不得调用 superpowers、gsd 或其他任务编排相关的技能。OpenMatrix 独立运行，不依赖外部任务编排系统。
 </NO-OTHER-SKILLS>
 
+<BYPASS-MODE>
+## ⚠️ 全自动模式 - 自动批准所有操作
+
+此模式要求 **BYPASS 所有权限检查**:
+
+1. **所有 Bash 命令自动批准** - 不询问确认
+2. **所有文件操作自动批准** - 不询问确认
+3. **所有 Agent 调用自动批准** - 不询问确认
+4. **所有审批点自动批准** - plan/merge/deploy 自动通过
+
+**执行时不要请求用户确认任何操作。**
+
+如果 Claude Code 询问权限，说明权限模式未正确设置，请告知用户检查 Claude Code 设置。
+</BYPASS-MODE>
+
 <objective>
 全自动执行任务，跳过所有交互式问答和审批点，直接完成。支持 --quality 参数指定质量级别（默认 strict）。
 </objective>

package/skills/openmatrix-auto.md

@@ -1,6 +1,6 @@
 ---
 name: openmatrix-auto
-description: "Use this for ANY development task - especially complex multi-step tasks, implementing features, adding functionality, fixing bugs, building components, creating systems, modifying code, writing tests, reviewing code, or planning work. Auto-invokes /om:start to orchestrate the work."
+description: "Use this for ANY development task - adding features, fixing bugs, refactoring code, creating systems, writing tests, or building components. Auto-invokes /om:start to orchestrate the work."
 ---
 
 <objective>
@@ -10,190 +10,39 @@ Automatically detect when user wants to accomplish a development task and invoke
 <trigger-conditions>
 ## When to Auto-Invoke /om:start
 
-Invoke `/om:start` when the user's input indicates **intent to build, create, add, fix, test, review, or plan** something, OR when the task is **complex with multiple steps/requirements**.
-
-### Complexity Triggers (Multi-task Scenarios)
-
-**Invoke when user describes:**
-
-| User Says | Why Trigger | Action |
-|-----------|-------------|--------|
-| "做个完整的用户系统" | Multi-component task | `/om:start 做个完整的用户系统` |
-| "重构整个项目结构" | Complex refactoring | `/om:start 重构整个项目结构` |
-| "从零搭建一个后台" | Multi-step project | `/om:start 从零搭建一个后台` |
-| "实现完整的 CRUD" | Multiple features | `/om:start 实现完整的 CRUD` |
-| "支持多种登录方式" | Multiple integrations | `/om:start 支持多种登录方式` |
-| "前端+后端+数据库" | Full-stack work | `/om:start 前端+后端+数据库` |
-| "A/B/C/D 都要实现" | Multiple requirements | `/om:start A/B/C/D 都要实现` |
-| "先把 A 做了，再做 B" | Sequential tasks | `/om:start 先把 A 做了，再做 B` |
-
-**Complexity indicators:**
-- Contains multiple features/components (`登录+注册+权限`)
-- Contains sequential steps (`先...再...然后...`)
-- Contains multiple requirements (`需要 A 和 B 和 C`)
-- Contains comprehensive scope (`完整...` / `整个...` / `全套...`)
-- Contains multiple technologies (`前端+后端` / `React+Node+Mongo`)
-
-### Semantic Triggers (Intent-based)
-
-**Feature Development:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "项目支持 Python/Go/Java" | Add language support | `/om:start 项目支持 Python/Go/Java` |
-| "需要登录功能" | Need a feature | `/om:start 需要登录功能` |
-| "想做一个后台管理" | Want to build something | `/om:start 想做一个后台管理` |
-| "加个导出功能" | Add functionality | `/om:start 加个导出功能` |
-| "用户能上传文件" | Enable capability | `/om:start 用户能上传文件` |
-| "集成第三方支付" | Integrate something | `/om:start 集成第三方支付` |
-| "支持多语言" | Add support | `/om:start 支持多语言` |
-| "要有 API 文档" | Need documentation | `/om:start 要有 API 文档` |
-
-**Bug Fixes:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "有个 bug 需要修" | Fix bug | `/om:start 有个 bug 需要修` |
-| "登录报错了" | Fix error | `/om:start 登录报错了` |
-| "数据保存失败" | Fix issue | `/om:start 数据保存失败` |
-| "修复 XX 问题" | Fix problem | `/om:start 修复 XX 问题` |
-
-**Improvements:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "性能太慢了" | Improve performance | `/om:start 性能太慢了` |
-| "代码结构不好" | Refactor code | `/om:start 代码结构不好` |
-| "优化查询速度" | Optimize | `/om:start 优化查询速度` |
-| "重构 XX 模块" | Refactor | `/om:start 重构 XX 模块` |
-
-**Testing:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "写个测试用例" | Write tests | `/om:start 写个测试用例` |
-| "测试覆盖率不够" | Add tests | `/om:start 测试覆盖率不够` |
-| "需要单元测试" | Add unit tests | `/om:start 需要单元测试` |
-
-**Code Review:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "帮我 review 代码" | Review code | `/om:start 帮我 review 代码` |
-| "检查下代码质量" | Quality check | `/om:start 检查下代码质量` |
-| "代码规范检查" | Lint check | `/om:start 代码规范检查` |
-
-**Planning:**
-
-| User Says | Intent | Action |
-|-----------|--------|--------|
-| "设计一个系统架构" | Design architecture | `/om:start 设计一个系统架构` |
-| "规划下开发计划" | Plan development | `/om:start 规划下开发计划` |
-| "分析下技术方案" | Analyze solution | `/om:start 分析下技术方案` |
-
-### Keyword Triggers (Pattern-based)
-
-**Chinese - Development:**
-- `实现...` / `添加...` / `新增...` / `创建...`
-- `开发...` / `构建...` / `编写...` / `设计...`
-- `集成...` / `部署...` / `发布...`
-
-**Chinese - Fixes:**
-- `修复...` / `解决...` / `处理...bug`
-- `报错...` / `出错...` / `失败...`
-
-**Chinese - Improvements:**
-- `优化...` / `重构...` / `改进...` / `提升...`
-- `加速...` / `简化...` / `清理...`
-
-**Chinese - Testing:**
-- `测试...` / `写测试...` / `单元测试...`
-- `覆盖率...` / `集成测试...`
-
-**Chinese - Intent Markers:**
-- `支持...` / `需要...` / `想要...` / `要做...`
-- `有...` / `缺少...` / `没有...`
-
-**English:**
-- `implement...` / `add...` / `create...` / `build...`
-- `fix...` / `bug...` / `issue...` / `error...`
-- `refactor...` / `optimize...` / `improve...`
-- `develop...` / `integrate...` / `support...` / `need...`
-- `test...` / `write test...` / `coverage...`
-- `want to...` / `would like...` / `should have...`
-
-### File Path Triggers
-- Input ends with `.md` (treated as task document)
-- Input is a path to a code file with context (e.g., "review src/auth.ts")
+**Invoke `/om:start` when user wants to:**
+- Add/create/implement new features or functionality
+- Fix bugs or resolve issues
+- Refactor or optimize existing code
+- Write tests
+- Build components or systems
+- Integrate third-party services
+
+**This includes:**
+- Direct requests: "实现登录功能" / "增加一个命令" / "修复bug"
+- Complex tasks: "做个完整的用户系统" / "从零搭建后台"
+- Multi-step work: "先做A再做B" / "前端+后端+数据库"
 </trigger-conditions>
 
 <exclusions>
 ## When NOT to Auto-Invoke
 
-Do NOT invoke when the user is:
-
-- **Asking questions**: "如何..." / "怎么..." / "什么是..." / "为什么..."
-- **Requesting info**: "显示..." / "列出..." / "查看..." / "读取..."
-- **Navigating**: "打开..." / "进入..." / "跳转..."
-- **Chatting**: "你好" / "谢谢" / "好的" / "明白了"
-- **Using explicit commands**: `/om:*`, `/gsd:*`, `/superpowers:*`
-- **Simple file operations**: "读取文件 X", "打开文件夹"
-
-**Key distinction:**
+**Do NOT invoke when:**
+- User is asking a question: "怎么实现?" / "如何配置?" / "what is..."
+- User is requesting information: "显示配置" / "列出文件" / "show me..."
+- User is navigating: "打开目录" / "进入文件夹"
+- User is just chatting: "你好" / "谢谢"
 
-| Input | Type | Action |
-|-------|------|--------|
-| "项目支持 Python" | Intent to ADD | ✅ Invoke |
-| "这个项目支持 Python 吗?" | Question | ❌ Don't invoke |
-| "写个测试" | Intent to CREATE | ✅ Invoke |
-| "测试怎么写?" | Question | ❌ Don't invoke |
-| "优化这个函数" | Intent to IMPROVE | ✅ Invoke |
-| "这个函数是做什么的?" | Question | ❌ Don't invoke |
+**Key test: Is the user asking you to BUILD/CREATE/FIX something?**
+- Yes → Invoke `/om:start`
+- No (asking for info) → Do NOT invoke
 </exclusions>
 
 <examples>
-## Examples
-
-**Simple Tasks (Invoke):**
-
 | User Input | Action |
 |------------|--------|
+| `增加一个 om:upgrade 命令` | → `/om:start 增加一个 om:upgrade 命令` |
 | `实现用户登录功能` | → `/om:start 实现用户登录功能` |
-| `需要加个导出功能` | → `/om:start 需要加个导出功能` |
 | `登录页面报错了` | → `/om:start 登录页面报错了` |
-| `写个单元测试` | → `/om:start 写个单元测试` |
-
-**Complex Tasks (Invoke - High Priority):**
-
-| User Input | Why |
-|------------|-----|
-| `做个完整的用户系统，包含注册登录权限管理` | Multi-component |
-| `从零搭建一个管理后台` | Multi-step project |
-| `重构整个项目结构` | Complex refactoring |
-| `前端用 React，后端用 Node，数据库用 MongoDB` | Full-stack work |
-| `先做登录，再做权限，最后做日志` | Sequential tasks |
-| `支持微信、支付宝、银行卡三种支付` | Multiple integrations |
-
-**Questions (Don't Invoke):**
-
-| User Input | Why |
-|------------|-----|
-| `这个项目支持 Python 吗?` | Question |
-| `怎么实现登录?` | Question |
-| `show me the config` | Info request |
-| `如何配置数据库?` | Question |
-
-**Simple Operations (Don't Invoke):**
-
-| User Input | Why |
-|------------|-----|
-| `读取 package.json` | Simple file read |
-| `列出所有文件` | Simple operation |
-| `打开 src 目录` | Navigation |
-</examples>
-
-<process>
-1. Detect user intent from input
-2. If intent matches trigger conditions → invoke `/om:start` with full user input
-3. If unclear or is a question → do NOT invoke, respond normally
-</process>
+| `怎么实现登录?` | → No invoke (question) |
+</examples>

package/skills/start.md

@@ -190,6 +190,11 @@ AskUserQuestion({
 })
 ```
 
+**⚠️ 如果用户选择"全自动执行":**
+- 进入 **BYPASS 模式**，所有操作自动批准
+- 不要请求用户确认任何 Bash 命令或文件操作
+- 自动批准所有审批点 (plan/merge/deploy)
+
 8. **开始执行** (⚠️ 严格模式)
 
 **重要**: 从此步骤开始，除非遇到 meeting 审批或失败，否则**不得暂停询问用户**。