@fitlab-ai/agent-infra 0.4.1 → 0.4.2

package/README.md

@@ -1,15 +1,25 @@
-# Agent Infra
+<p align="center">
+  <img src="./assets/logo.svg" alt="Agent Infra Logo" width="200">
+</p>
 
-[![npm version](https://img.shields.io/npm/v/@fitlab-ai/agent-infra)](https://www.npmjs.com/package/@fitlab-ai/agent-infra)
-[![npm downloads](https://img.shields.io/npm/dm/@fitlab-ai/agent-infra)](https://www.npmjs.com/package/@fitlab-ai/agent-infra)
-[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](License.txt)
-[![Node.js >= 18](https://img.shields.io/badge/Node.js-%3E%3D18-brightgreen?logo=node.js)](https://nodejs.org/)
-[![GitHub release](https://img.shields.io/github/v/release/fitlab-ai/agent-infra)](https://github.com/fitlab-ai/agent-infra/releases)
-[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](CONTRIBUTING.md)
+<h1 align="center">Agent Infra</h1>
 
-The missing collaboration layer for AI coding agents — unified skills and workflows for Claude Code, Codex, Gemini CLI, and OpenCode.
+<p align="center">
+  The missing collaboration layer for AI coding agents — unified skills and workflows for Claude Code, Codex, Gemini CLI, and OpenCode.
+</p>
 
-**Semi-automated programming, powered by AI agents.** Define a requirement, let AI handle analysis, planning, coding, review, and delivery — you only step in when it matters.
+<p align="center">
+  <strong>Semi-automated programming, powered by AI agents.</strong> Define a requirement, let AI handle analysis, planning, coding, review, and delivery — you only step in when it matters.
+</p>
+
+<p align="center">
+  <a href="https://www.npmjs.com/package/@fitlab-ai/agent-infra"><img src="https://img.shields.io/npm/v/@fitlab-ai/agent-infra" alt="npm version"></a>
+  <a href="https://www.npmjs.com/package/@fitlab-ai/agent-infra"><img src="https://img.shields.io/npm/dm/@fitlab-ai/agent-infra" alt="npm downloads"></a>
+  <a href="License.txt"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT"></a>
+  <a href="https://nodejs.org/"><img src="https://img.shields.io/badge/Node.js-%3E%3D18-brightgreen?logo=node.js" alt="Node.js >= 18"></a>
+  <a href="https://github.com/fitlab-ai/agent-infra/releases"><img src="https://img.shields.io/github/v/release/fitlab-ai/agent-infra" alt="GitHub release"></a>
+  <a href="CONTRIBUTING.md"><img src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg" alt="PRs Welcome"></a>
+</p>
 
 [中文版](README.zh-CN.md)
 
@@ -74,7 +84,7 @@ Just fix it at the application layer in LoginService.
 
 ```bash
 /commit
-/create-pr
+/create-pr TASK-20260319-100000
 /complete-task TASK-20260319-100000
 ```
 
@@ -264,8 +274,8 @@ agent-infra ships with **28 built-in AI skills**. They are organized by use case
 |-------|-------------|------------|----------------------|
 | `create-issue` | Create a GitHub Issue from a task file. | `task-id` | Push a local task into GitHub tracking. |
 | `sync-issue` | Post task progress updates back to the linked GitHub Issue. | `task-id` or `issue-number` | Keep stakeholders updated as the task evolves. |
-| `create-pr` | Open a Pull Request to an inferred or explicit target branch. | `target-branch` (optional) | Publish reviewed changes for merge. |
-| `sync-pr` | Sync task progress and review metadata into the Pull Request. | `task-id` | Keep PR metadata aligned with the local task record. |
+| `create-pr` | Open a Pull Request to an inferred or explicit target branch. | `task-id` (optional), `target-branch` (optional) | Publish reviewed changes for merge, with optional explicit task linkage after a fresh session. |
+| `sync-pr` | Sync task progress and review metadata into the Pull Request. | `task-id` or `pr-number` | Keep PR metadata aligned with the local task record from either task or PR entrypoints. |
 
 <a id="code-quality"></a>
 
@@ -380,7 +390,7 @@ The generated `.agents/.airc.json` file is the central contract between the boot
   "project": "my-project",
   "org": "my-org",
   "language": "en",
-  "templateVersion": "v0.4.1",
+  "templateVersion": "v0.4.2",
   "files": {
     "managed": [
       ".agents/workspace/README.md",

package/README.zh-CN.md

@@ -1,15 +1,25 @@
-# Agent Infra
+<p align="center">
+  <img src="./assets/logo.svg" alt="Agent Infra 标志" width="200">
+</p>
 
-[![npm version](https://img.shields.io/npm/v/@fitlab-ai/agent-infra)](https://www.npmjs.com/package/@fitlab-ai/agent-infra)
-[![npm downloads](https://img.shields.io/npm/dm/@fitlab-ai/agent-infra)](https://www.npmjs.com/package/@fitlab-ai/agent-infra)
-[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](License.txt)
-[![Node.js >= 18](https://img.shields.io/badge/Node.js-%3E%3D18-brightgreen?logo=node.js)](https://nodejs.org/)
-[![GitHub release](https://img.shields.io/github/v/release/fitlab-ai/agent-infra)](https://github.com/fitlab-ai/agent-infra/releases)
-[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](CONTRIBUTING.md)
+<h1 align="center">Agent Infra</h1>
 
-AI 编程代理之间缺失的协作层 —— 为 Claude Code、Codex、Gemini CLI、OpenCode 提供统一的 skills 和工作流。
+<p align="center">
+  AI 编程代理之间缺失的协作层 —— 为 Claude Code、Codex、Gemini CLI、OpenCode 提供统一的 skills 和工作流。
+</p>
 
-**AI Agent 半自动化编程神器。** 定义需求，让 AI 完成分析、方案设计、编码、审查和交付 —— 你只需在关键节点介入。
+<p align="center">
+  <strong>AI Agent 半自动化编程神器。</strong> 定义需求，让 AI 完成分析、方案设计、编码、审查和交付 —— 你只需在关键节点介入。
+</p>
+
+<p align="center">
+  <a href="https://www.npmjs.com/package/@fitlab-ai/agent-infra"><img src="https://img.shields.io/npm/v/@fitlab-ai/agent-infra" alt="npm version"></a>
+  <a href="https://www.npmjs.com/package/@fitlab-ai/agent-infra"><img src="https://img.shields.io/npm/dm/@fitlab-ai/agent-infra" alt="npm downloads"></a>
+  <a href="License.txt"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT"></a>
+  <a href="https://nodejs.org/"><img src="https://img.shields.io/badge/Node.js-%3E%3D18-brightgreen?logo=node.js" alt="Node.js >= 18"></a>
+  <a href="https://github.com/fitlab-ai/agent-infra/releases"><img src="https://img.shields.io/github/v/release/fitlab-ai/agent-infra" alt="GitHub release"></a>
+  <a href="CONTRIBUTING.md"><img src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg" alt="PRs Welcome"></a>
+</p>
 
 [English](README.md)
 
@@ -74,7 +84,7 @@ agent-infra 的目标就是把这层协作面标准化。它为所有支持的 A
 
 ```bash
 /commit
-/create-pr
+/create-pr TASK-20260319-100000
 /complete-task TASK-20260319-100000
 ```
 
@@ -264,8 +274,8 @@ agent-infra 内置 **28 个 AI skills**。它们按使用场景分组，但共
 |-------|------|------|---------|
 | `create-issue` | 根据任务文件创建 GitHub Issue。 | `task-id` | 需要把本地任务同步到 GitHub 跟踪时。 |
 | `sync-issue` | 将任务进度同步回关联的 GitHub Issue。 | `task-id` 或 `issue-number` | 任务推进过程中持续同步状态。 |
-| `create-pr` | 向推断出的目标分支或显式指定分支创建 Pull Request。 | `target-branch`（可选） | 变更准备合入时创建 PR。 |
-| `sync-pr` | 将任务进度和审查元数据同步到 Pull Request。 | `task-id` | 让 PR 元数据与本地任务记录保持一致。 |
+| `create-pr` | 向推断出的目标分支或显式指定分支创建 Pull Request。 | `task-id`（可选）、`target-branch`（可选） | 变更准备合入时创建 PR；清空上下文后也可显式传入任务关联。 |
+| `sync-pr` | 将任务进度和审查元数据同步到 Pull Request。 | `task-id` 或 `pr-number` | 允许从任务入口或 PR 入口同步，让 PR 元数据与本地任务记录保持一致。 |
 
 <a id="code-quality"></a>
 
@@ -380,7 +390,7 @@ import-issue #42                    从 GitHub Issue 导入任务
   "project": "my-project",
   "org": "my-org",
   "language": "en",
-  "templateVersion": "v0.4.1",
+  "templateVersion": "v0.4.2",
   "files": {
     "managed": [
       ".agents/workspace/README.md",

package/lib/defaults.json

@@ -1,6 +1,8 @@
 {
   "files": {
     "managed": [
+      ".agents/QUICKSTART.md",
+      ".agents/README.md",
       ".agents/skills/",
       ".agents/templates/",
       ".agents/workflows/",
@@ -16,22 +18,14 @@
       "**/test-integration.*",
       "**/test.*",
       "**/upgrade-dependency.*",
-      ".agents/QUICKSTART.md",
-      ".agents/README.md",
       ".agents/skills/release/SKILL.*",
       ".agents/skills/test-integration/SKILL.*",
       ".agents/skills/test/SKILL.*",
       ".agents/skills/upgrade-dependency/SKILL.*",
-      ".claude/CLAUDE.md",
-      ".claude/project-rules.md",
       ".claude/settings.json",
-      ".codex/README.md",
       ".gemini/settings.json",
       ".github/hooks/pre-commit",
-      ".gitignore",
-      ".opencode/COMMAND_STYLE_GUIDE.md",
-      ".opencode/README.md",
-      "AGENTS.md"
+      ".gitignore"
     ],
     "ejected": []
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@fitlab-ai/agent-infra",
-  "version": "0.4.1",
+  "version": "0.4.2",
   "description": "Bootstrap tool for AI multi-tool collaboration infrastructure — works with Claude Code, Codex, Gemini CLI, and OpenCode",
   "license": "MIT",
   "type": "module",

package/templates/.agents/QUICKSTART.md

@@ -4,7 +4,7 @@ This guide walks you through using multiple AI coding assistants together on a p
 
 ## Prerequisites
 
-- At least one AI coding tool installed (Claude Code, Codex CLI, Gemini CLI, or Cursor)
+- At least one AI coding tool installed (Claude Code, Codex CLI, Gemini CLI, or OpenCode)
 - A project with `.agents/` directory set up (this project)
 - Familiarity with your project's codebase
 
@@ -32,7 +32,7 @@ cp .agents/templates/task.md .agents/workspace/active/task-001.md
 id: task-001
 type: feature          # feature | bugfix | refactor | docs | review
 status: open           # open | in-progress | review | blocked | completed
-assigned_to: claude    # claude | codex | gemini | cursor | human
+assigned_to: claude    # claude | codex | gemini | opencode | human
 ```
 
 3. Describe the task in the body of the document.
@@ -63,7 +63,7 @@ gemini
 #  create a technical design. Define interfaces and outline the approach."
 ```
 
-### Phase 3: Implementation (Recommended: Codex CLI or Cursor)
+### Phase 3: Implementation (Recommended: Codex CLI or OpenCode)
 
 ```bash
 # Switch to Codex CLI for implementation
@@ -91,7 +91,7 @@ claude
 ### Bug Fix
 
 1. **Reproduce & Analyze** (Claude Code): Identify the root cause.
-2. **Implement Fix** (Codex CLI / Cursor): Write the fix and tests.
+2. **Implement Fix** (Codex CLI / OpenCode): Write the fix and tests.
 3. **Review** (Claude Code): Verify the fix is correct and complete.
 4. **Commit**: Create PR with bug fix description.
 
@@ -100,7 +100,7 @@ claude
 cp .agents/templates/task.md .agents/workspace/active/bugfix-001.md
 # Edit the task, then:
 # 1. Use Claude Code to analyze
-# 2. Use Codex/Cursor to fix
+# 2. Use Codex/OpenCode to fix
 # 3. Use Claude Code to review
 ```
 
@@ -119,7 +119,7 @@ cp .agents/templates/review-report.md .agents/workspace/active/review-pr-42.md
 
 1. **Analyze Scope** (Claude Code / Gemini CLI): Map all affected areas.
 2. **Design** (Claude Code): Plan the refactoring approach.
-3. **Implement** (Codex CLI / Cursor): Execute the refactoring.
+3. **Implement** (Codex CLI / OpenCode): Execute the refactoring.
 4. **Verify** (Claude Code): Ensure no regressions, run tests.
 
 ```bash
@@ -156,7 +156,7 @@ Even if you're switching between AIs quickly, a brief handoff note saves time an
 ### 3. Use the Right Tool for the Job
 
 - Complex analysis? Use Claude Code or Gemini CLI.
-- Straightforward implementation? Use Codex CLI or Cursor.
+- Straightforward implementation? Use Codex CLI or OpenCode.
 - Large file review? Use Gemini CLI.
 
 ### 4. Keep Tasks Small

package/templates/.agents/QUICKSTART.zh-CN.md

@@ -4,7 +4,7 @@
 
 ## 前提条件
 
-- 至少安装一个 AI 编程工具（Claude Code、Codex CLI、Gemini CLI 或 Cursor）
+- 至少安装一个 AI 编程工具（Claude Code、Codex CLI、Gemini CLI 或 OpenCode）
 - 项目已设置 `.agents/` 目录（本项目已就绪）
 - 熟悉你的项目代码库
 
@@ -23,7 +23,7 @@ git config core.hooksPath .github/hooks
 1. 将任务模板复制到活跃工作区：
 
 ```bash
-cp .agents/templates/task.zh-CN.md .agents/workspace/active/task-001.md
+cp .agents/templates/task.md .agents/workspace/active/task-001.md
 ```
 
 2. 填写任务元数据：
@@ -32,7 +32,7 @@ cp .agents/templates/task.zh-CN.md .agents/workspace/active/task-001.md
 id: task-001
 type: feature          # feature | bugfix | refactor | docs | review
 status: open           # open | in-progress | review | blocked | completed
-assigned_to: claude    # claude | codex | gemini | cursor | human
+assigned_to: claude    # claude | codex | gemini | opencode | human
 ```
 
 3. 在文档正文中描述任务。
@@ -63,7 +63,7 @@ gemini
 #  创建技术设计方案。定义接口并概述实现思路。"
 ```
 
-### 阶段 3：实现（推荐：Codex CLI 或 Cursor）
+### 阶段 3：实现（推荐：Codex CLI 或 OpenCode）
 
 ```bash
 # 切换到 Codex CLI 进行实现
@@ -91,16 +91,16 @@ claude
 ### 缺陷修复
 
 1. **复现和分析**（Claude Code）：识别根本原因。
-2. **实现修复**（Codex CLI / Cursor）：编写修复代码和测试。
+2. **实现修复**（Codex CLI / OpenCode）：编写修复代码和测试。
 3. **审查**（Claude Code）：验证修复是否正确和完整。
 4. **提交**：创建包含缺陷修复描述的 PR。
 
 ```bash
 # 快速缺陷修复工作流
-cp .agents/templates/task.zh-CN.md .agents/workspace/active/bugfix-001.md
+cp .agents/templates/task.md .agents/workspace/active/bugfix-001.md
 # 编辑任务，然后：
 # 1. 使用 Claude Code 分析
-# 2. 使用 Codex/Cursor 修复
+# 2. 使用 Codex/OpenCode 修复
 # 3. 使用 Claude Code 审查
 ```
 
@@ -111,7 +111,7 @@ cp .agents/templates/task.zh-CN.md .agents/workspace/active/bugfix-001.md
 3. **报告**：从模板生成审查报告。
 
 ```bash
-cp .agents/templates/review-report.zh-CN.md .agents/workspace/active/review-pr-42.md
+cp .agents/templates/review-report.md .agents/workspace/active/review-pr-42.md
 # 使用 Claude Code 填写审查内容
 ```
 
@@ -119,12 +119,12 @@ cp .agents/templates/review-report.zh-CN.md .agents/workspace/active/review-pr-4
 
 1. **分析范围**（Claude Code / Gemini CLI）：映射所有受影响区域。
 2. **设计**（Claude Code）：规划重构方案。
-3. **实现**（Codex CLI / Cursor）：执行重构。
+3. **实现**（Codex CLI / OpenCode）：执行重构。
 4. **验证**（Claude Code）：确保没有回归问题，运行测试。
 
 ```bash
-cp .agents/templates/task.zh-CN.md .agents/workspace/active/refactor-001.md
-# 按照 .agents/workflows/refactoring.zh-CN.yaml 中的重构工作流执行
+cp .agents/templates/task.md .agents/workspace/active/refactor-001.md
+# 按照 .agents/workflows/refactoring.yaml 中的重构工作流执行
 ```
 
 ## 创建交接文档
@@ -132,7 +132,7 @@ cp .agents/templates/task.zh-CN.md .agents/workspace/active/refactor-001.md
 在 AI 工具之间切换时，创建交接文档：
 
 ```bash
-cp .agents/templates/handoff.zh-CN.md .agents/workspace/active/handoff-task-001-phase2.md
+cp .agents/templates/handoff.md .agents/workspace/active/handoff-task-001-phase2.md
 ```
 
 填写：
@@ -156,7 +156,7 @@ cp .agents/templates/handoff.zh-CN.md .agents/workspace/active/handoff-task-001-
 ### 3. 选择合适的工具
 
 - 复杂分析？使用 Claude Code 或 Gemini CLI。
-- 直接的实现工作？使用 Codex CLI 或 Cursor。
+- 直接的实现工作？使用 Codex CLI 或 OpenCode。
 - 大文件审查？使用 Gemini CLI。
 
 ### 4. 保持任务小而精

package/templates/.agents/README.md

@@ -1,6 +1,6 @@
 # Multi-AI Collaboration Guide
 
-This project supports collaboration across multiple AI coding assistants, including Claude Code, OpenAI Codex CLI, Gemini CLI, Cursor, and others.
+This project supports collaboration across multiple AI coding assistants, including Claude Code, OpenAI Codex CLI, Gemini CLI, OpenCode, and others.
 
 ## Dual-Config Architecture
 
@@ -8,10 +8,10 @@ Different AI tools read configuration from different locations:
 
 | AI Tool | Primary Config | Fallback |
 |---------|---------------|----------|
-| Claude Code | `.claude/` (CLAUDE.md, commands/, settings/) | - |
+| Claude Code | `.claude/` (CLAUDE.md, commands/, settings.json) | - |
 | OpenAI Codex CLI | `AGENTS.md` | - |
 | Gemini CLI | `AGENTS.md` | - |
-| Cursor | `.cursorrules` or `.cursor/rules/` | `AGENTS.md` |
+| OpenCode | `AGENTS.md` | - |
 | Other AI Tools | `AGENTS.md` | Project README |
 
 - **Claude Code** uses its dedicated `.claude/` directory for project instructions, slash commands, and settings.
@@ -23,10 +23,8 @@ This dual-config approach ensures every AI tool receives appropriate project con
 
 ```
 .agents/                        # AI collaboration config (version-controlled)
-  README.md                     # This file (English)
-  README.zh-CN.md               # Chinese version
+  README.md                     # Collaboration guide
   QUICKSTART.md                 # Quick start guide
-  QUICKSTART.zh-CN.md           # Quick start guide (Chinese)
   templates/                    # Task and document templates
     task.md                     # Task template
     handoff.md                  # AI-to-AI handoff template
@@ -45,7 +43,7 @@ This dual-config approach ensures every AI tool receives appropriate project con
 .claude/                        # Claude Code specific config
   CLAUDE.md                     # Project instructions for Claude
   commands/                     # Slash commands
-  settings/                     # Claude settings
+  settings.json                 # Claude settings
 ```
 
 ## Collaboration Model
@@ -76,21 +74,21 @@ When one AI completes a phase, it produces a **handoff document** (see `.agents/
 
 Each AI tool has different strengths. Use them accordingly:
 
-| Capability | Claude Code | Codex CLI | Gemini CLI | Cursor |
-|-----------|-------------|-----------|------------|--------|
+| Capability | Claude Code | Codex CLI | Gemini CLI | OpenCode |
+|-----------|-------------|-----------|------------|----------|
 | Codebase analysis | Excellent | Good | Excellent | Good |
-| Code review | Excellent | Good | Good | Fair |
+| Code review | Excellent | Good | Good | Good |
 | Implementation | Good | Excellent | Good | Excellent |
-| Large context | Good | Fair | Excellent | Good |
-| Refactoring | Good | Good | Good | Excellent |
-| Documentation | Excellent | Good | Good | Fair |
+| Large context | Good | Fair | Excellent | Fair |
+| Refactoring | Good | Good | Good | Good |
+| Documentation | Excellent | Good | Good | Good |
 
 ### Recommended Assignments
 
 - **Analysis & Review** - Claude Code (strong reasoning, thorough exploration)
-- **Implementation** - Codex CLI or Cursor (fast code generation, inline editing)
+- **Implementation** - Codex CLI or OpenCode (fast code generation, command-driven editing)
 - **Large Context Tasks** - Gemini CLI (large context window for cross-file analysis)
-- **Quick Edits** - Cursor (IDE integration, fast iteration)
+- **Command-Driven Iteration** - OpenCode (workflow-friendly TUI execution)
 
 ## Quick Start
 
@@ -118,10 +116,11 @@ When writing or updating `.agents/skills/*/SKILL.md` files and their templates,
 
 1. Use consecutive integers for top-level steps: `1.`, `2.`, `3.`.
 2. Use nested numbering only for child actions that belong to a parent step: `1.1`, `1.2`, `2.1`.
-3. Use `a`, `b`, and `c` markers for branches, conditions, or alternative paths within the same step.
+3. Use `a`, `b`, and `c` markers for branches, conditions, or alternative paths within the same step; keep them scoped to child options rather than standalone decision tracks or output templates.
 4. Do not use intermediate numbers such as `1.5` or `2.5`; if a new standalone step is needed, renumber the following top-level steps.
 5. When renumbering, update every in-document step reference so the instructions remain accurate.
 6. Extract long bash scripts into a sibling `scripts/` directory; the SKILL.md should contain only a single-line invocation (e.g., `bash .agents/skills/<skill>/scripts/<script>.sh`) and a brief summary of the script's responsibilities.
+7. In SKILL.md files and their `reference/` templates, use “Scenario” naming for standalone condition branches, decision paths, or output templates (for example, “Scenario A”).
 
 ### SKILL.md Size Control
 

package/templates/.agents/README.zh-CN.md

@@ -1,6 +1,6 @@
 # 多 AI 协作指南
 
-本项目支持多个 AI 编程助手协同工作，包括 Claude Code、OpenAI Codex CLI、Gemini CLI、Cursor 等。
+本项目支持多个 AI 编程助手协同工作，包括 Claude Code、OpenAI Codex CLI、Gemini CLI、OpenCode 等。
 
 ## 双配置架构
 
@@ -8,10 +8,10 @@
 
 | AI 工具 | 主要配置 | 备选配置 |
 |---------|---------|---------|
-| Claude Code | `.claude/`（CLAUDE.md、commands/、settings/） | - |
+| Claude Code | `.claude/`（CLAUDE.md、commands/、settings.json） | - |
 | OpenAI Codex CLI | `AGENTS.md` | - |
 | Gemini CLI | `AGENTS.md` | - |
-| Cursor | `.cursorrules` 或 `.cursor/rules/` | `AGENTS.md` |
+| OpenCode | `AGENTS.md` | - |
 | 其他 AI 工具 | `AGENTS.md` | 项目 README |
 
 - **Claude Code** 使用专属的 `.claude/` 目录存放项目指令、斜杠命令和设置。
@@ -23,10 +23,8 @@
 
 ```
 .agents/                        # AI 协作配置（版本控制）
-  README.md                     # 本文件（英文）
-  README.zh-CN.md               # 中文版本
+  README.md                     # 协作指南
   QUICKSTART.md                 # 快速入门指南
-  QUICKSTART.zh-CN.md           # 快速入门指南（中文）
   templates/                    # 任务和文档模板
     task.md                     # 任务模板
     handoff.md                  # AI 间交接模板
@@ -45,7 +43,7 @@
 .claude/                        # Claude Code 专属配置
   CLAUDE.md                     # Claude 项目指令
   commands/                     # 斜杠命令
-  settings/                     # Claude 设置
+  settings.json                 # Claude 设置
 ```
 
 ## 协作模型
@@ -76,26 +74,26 @@
 
 每个 AI 工具有不同的优势，请据此分配任务：
 
-| 能力 | Claude Code | Codex CLI | Gemini CLI | Cursor |
-|-----|-------------|-----------|------------|--------|
+| 能力 | Claude Code | Codex CLI | Gemini CLI | OpenCode |
+|-----|-------------|-----------|------------|----------|
 | 代码库分析 | 优秀 | 良好 | 优秀 | 良好 |
-| 代码审查 | 优秀 | 良好 | 良好 | 一般 |
+| 代码审查 | 优秀 | 良好 | 良好 | 良好 |
 | 代码实现 | 良好 | 优秀 | 良好 | 优秀 |
-| 大上下文处理 | 良好 | 一般 | 优秀 | 良好 |
-| 重构 | 良好 | 良好 | 良好 | 优秀 |
-| 文档编写 | 优秀 | 良好 | 良好 | 一般 |
+| 大上下文处理 | 良好 | 一般 | 优秀 | 一般 |
+| 重构 | 良好 | 良好 | 良好 | 良好 |
+| 文档编写 | 优秀 | 良好 | 良好 | 良好 |
 
 ### 推荐分配
 
 - **分析和审查** - Claude Code（推理能力强，探索全面）
-- **代码实现** - Codex CLI 或 Cursor（代码生成快，内联编辑）
+- **代码实现** - Codex CLI 或 OpenCode（代码生成快，命令式工作流顺手）
 - **大上下文任务** - Gemini CLI（大上下文窗口，适合跨文件分析）
-- **快速编辑** - Cursor（IDE 集成，快速迭代）
+- **命令式迭代** - OpenCode（适合按工作流连续推进）
 
 ## 快速入门
 
-1. **阅读快速入门指南**：参见 `QUICKSTART.zh-CN.md` 获取分步说明。
-2. **创建任务**：将 `.agents/templates/task.zh-CN.md` 复制到 `.agents/workspace/active/`。
+1. **阅读快速入门指南**：参见 `QUICKSTART.md` 获取分步说明。
+2. **创建任务**：将 `.agents/templates/task.md` 复制到 `.agents/workspace/active/`。
 3. **分配给 AI**：更新任务元数据中的 `assigned_to` 字段。
 4. **执行工作流**：按照 `.agents/workflows/` 中相应的工作流执行。
 5. **交接**：切换 AI 时，从模板创建交接文档。
@@ -118,10 +116,11 @@
 
 1. 顶级步骤使用连续整数：`1.`、`2.`、`3.`。
 2. 只有父步骤下的从属动作才使用子步骤：`1.1`、`1.2`、`2.1`。
-3. 同一步中的分支、条件或多种可能性使用 `a`、`b`、`c` 标记。
+3. 同一步中的从属选项、条件分支或并列可能性使用 `a`、`b`、`c` 标记；仅用于步骤内部的子项展开，不用于命名独立的决策路径或输出模板。
 4. 不要使用 `1.5`、`2.5` 这类中间编号；如新增独立步骤，应整体顺延后续编号。
 5. 调整编号时，必须同步更新文中的步骤引用，确保说明、命令和检查点一致。
 6. 长 bash 脚本应从 SKILL.md 提取到同级 `scripts/` 目录中，SKILL.md 只保留单行调用（如 `bash .agents/skills/<skill>/scripts/<script>.sh`）和对脚本职责的概要说明。
+7. 在 SKILL.md 及其 `reference/` 模板中，如需为独立的条件分流、决策路径或输出模板命名，统一使用“场景”命名（例如使用“场景 A”）。
 
 ### SKILL.md 体积控制
 

package/templates/.agents/skills/commit/reference/task-status-update.md

@@ -82,7 +82,7 @@ Required next-step commands:
 
 ```text
 Next step - create Pull Request:
-  - Claude Code / OpenCode: /create-pr
-  - Gemini CLI: /agent-infra:create-pr
-  - Codex CLI: $create-pr
+  - Claude Code / OpenCode: /create-pr {task-id}
+  - Gemini CLI: /agent-infra:create-pr {task-id}
+  - Codex CLI: $create-pr {task-id}
 ```

package/templates/.agents/skills/commit/reference/task-status-update.zh-CN.md

@@ -82,7 +82,7 @@ date "+%Y-%m-%d %H:%M:%S"
 
 ```text
 下一步 - 创建 Pull Request：
-  - Claude Code / OpenCode: /create-pr
-  - Gemini CLI: /agent-infra:create-pr
-  - Codex CLI: $create-pr
+  - Claude Code / OpenCode: /create-pr {task-id}
+  - Gemini CLI: /agent-infra:create-pr {task-id}
+  - Codex CLI: $create-pr {task-id}
 ```

package/templates/.agents/skills/create-pr/SKILL.md

@@ -9,38 +9,47 @@ Create a Pull Request and, when task-related, sync the essential metadata immedi
 
 ## Execution Flow
 
-### 1. Determine the Target Branch
+### 1. Parse Command Arguments
+
+Identify arguments from the command input:
+- arguments matching `TASK-{yyyyMMdd-HHmmss}` -> `{task-id}`
+- remaining arguments -> `{target-branch}`
+
+If `{task-id}` is provided, read `.agents/workspace/active/{task-id}/task.md` to load task metadata such as `issue_number` and `type`.
+If `{task-id}` is omitted, try to resolve it from the current session context; if it still cannot be determined, skip task-association logic in later steps.
+
+### 2. Determine the Target Branch
 
 Use the explicit argument when provided. Otherwise infer the target branch from Git history and branch topology.
 
 > Detailed branch detection rules live in `reference/branch-strategy.md`. Read `reference/branch-strategy.md` before auto-detecting the base branch.
 
-### 2. Prepare the PR Body
+### 3. Prepare the PR Body
 
 Read `.github/PULL_REQUEST_TEMPLATE.md` when it exists, review recent merged PRs for style, and gather all commits between `<target-branch>` and `HEAD`.
 
 > Template handling, HEREDOC body generation, and `Generated with AI assistance` requirements live in `reference/pr-body-template.md`. Read `reference/pr-body-template.md` before writing the PR body.
 
-### 3. Check Remote Branch State
+### 4. Check Remote Branch State
 
 Confirm whether the current branch already has an upstream. Push with `git push -u origin <current-branch>` when required.
 
-### 4. Create the PR
+### 5. Create the PR
 
 Create the PR with `gh pr create --base <target-branch> --title "<title>" --assignee @me --body ...`.
 
-If a related task provides `issue_number`, keep `Closes #{issue-number}` in the PR body.
+If `{task-id}` is available and the related task provides `issue_number`, keep `Closes #{issue-number}` in the PR body.
 
-### 5. Sync PR Metadata
+### 6. Sync PR Metadata
 
-For task-related PRs, sync the core metadata immediately:
+For PRs where `{task-id}` is available, sync the core metadata immediately:
 - run `gh label list --search "type:" --limit 1 --json name --jq 'length'`
 - add the mapped type label with `gh pr edit {pr-number} --add-label "{type-label}"`
 - add relevant `in: {module}` labels with `gh pr edit {pr-number} --add-label "in: {module}"`
 - set the milestone with `gh pr edit {pr-number} --milestone "{milestone-title}"`
 - keep Development linking in the PR body with `Closes #{issue-number}` when applicable
 
-### 6. Update Task Status
+### 7. Update Task Status
 
 Get the current time:
 
@@ -48,9 +57,9 @@ Get the current time:
 date "+%Y-%m-%d %H:%M:%S"
 ```
 
-If this PR is task-related, update task.md with `pr_number`, `updated_at`, and append the PR Created Activity Log entry.
+If `{task-id}` is available, update task.md with `pr_number`, `updated_at`, and append the PR Created Activity Log entry.
 
-### 7. Inform the User
+### 8. Inform the User
 
 Explain the created PR URL, summarize metadata sync results, and present both follow-up commands in order:
 - optional `sync-pr {task-id}` to publish reviewer-facing context
@@ -59,7 +68,7 @@ Explain the created PR URL, summarize metadata sync results, and present both fo
 ## Notes
 
 - Review every commit in the branch, not only the latest one
-- `create-pr` must not defer type-label mapping to `sync-pr`; inline the mapping here
+- `create-pr` must not defer type-label mapping to `sync-pr`; inline the mapping here when `{task-id}` is available
 - When metadata inheritance from the Issue fails, continue with task.md and branch-based fallbacks
 
 ## Error Handling