npm - @pzy560117/codex-harness - Versions diffs - 0.1.4 → 0.1.5 - Mend

@pzy560117/codex-harness 0.1.4 → 0.1.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (463) hide show

package/package-source/docs/codex-harness-engineering/templates/package-assets/skills/git-workflow/SKILL.md ADDED Viewed

@@ -0,0 +1,108 @@
+---
+name: git-workflow
+description: Git Workflow Skill
+---
+# Git Workflow Skill
+Git 版本控制最佳实践和工作流规范。
+## 分支策略
+### Git Flow 模型
+- **main/master**: 生产环境代码，始终可部署
+- **develop**: 开发主分支，集成最新功能
+- **feature/xxx**: 功能分支，从 develop 创建
+- **hotfix/xxx**: 紧急修复，从 main 创建
+- **release/xxx**: 发布准备分支
+### 分支命名规范
+```
+feature/user-authentication
+feature/JIRA-123-add-login
+bugfix/fix-null-pointer
+hotfix/security-patch-v1.2.1
+release/v2.0.0
+```
+## Commit Message 规范
+### Conventional Commits 格式
+```
+<type>(<scope>): <subject>
+<body>
+<footer>
+```
+### Type 类型
+| Type | 说明 |
+|------|------|
+| feat | 新功能 |
+| fix | Bug 修复 |
+| docs | 文档更新 |
+| style | 代码格式（不影响逻辑） |
+| refactor | 重构（非新功能、非修复） |
+| perf | 性能优化 |
+| test | 测试相关 |
+| chore | 构建/工具变动 |
+| ci | CI 配置变更 |
+### 示例
+```
+feat(auth): 添加 OAuth2.0 登录支持
+- 集成 Google OAuth
+- 添加 token 刷新机制
+- 更新用户模型
+Closes #123
+```
+## 操作规范
+### 合并策略
+- **功能分支**: Squash merge 保持历史整洁
+- **发布分支**: Merge commit 保留完整历史
+- **Hotfix**: Cherry-pick 或直接 merge
+### 禁止事项
+- 不在 main/develop 上直接提交
+- 不 force push 公共分支
+- 不提交敏感信息（密钥、密码）
+- 不提交大型二进制文件
+### .gitignore 必备项
+```
+node_modules/
+.env
+.env.local
+*.log
+.DS_Store
+dist/
+build/
+coverage/
+.idea/
+.vscode/settings.json
+```
+## 常用命令
+```bash
+# 交互式 rebase 整理提交
+git rebase -i HEAD~3
+# 暂存当前修改
+git stash push -m "WIP: feature description"
+# 查看文件修改历史
+git log --follow -p -- <file>
+# 撤销最近一次提交（保留修改）
+git reset --soft HEAD~1
+# 修改最近一次提交信息
+git commit --amend -m "new message"
+```

package/package-source/docs/codex-harness-engineering/templates/package-assets/skills/git-xianyu-analyzer/SKILL.md ADDED Viewed

@@ -0,0 +1,346 @@
+---
+name: git-xianyu-analyzer
+description: Automates cloning a git repository, removing git history, analyzing the project comprehensively, and generating a Xianyu sales post. Use when the user provides a git URL and wants to prepare a project for Xianyu listing or sharing.
+---
+# Git Project Analyzer for Xianyu
+Automates the workflow of cloning, cleaning, analyzing a project, and generating a professional Xianyu sales post.
+## Workflow
+### 1. Clone Repository
+- Execute `git clone <url>` in the current working directory
+- Git will automatically create a directory based on the repository name
+- Verify clone success by checking if the directory exists
+### 2. Clean Git History
+- Remove the `.git` directory to detach from remote tracking
+- Windows: `Remove-Item -Recurse -Force <repo_name>\.git`
+- Unix/Linux/Mac: `rm -rf <repo_name>/.git`
+### 3. Clean Open Source Information and Modify Documentation
+**重要：隐藏开源来源，使项目看起来像独立开发的商业项目**
+#### 3.1 删除开源标识文件
+- Delete `LICENSE` or `LICENSE.txt` file
+- Delete `.github` directory (contains GitHub workflows, issue templates)
+- Delete `CONTRIBUTING.md`, `CODE_OF_CONDUCT.md` if exists
+- Delete `CHANGELOG.md` or `HISTORY.md` if contains git history references
+#### 3.2 修改 README.md
+**必须移除的内容：**
+- ❌ MIT/Apache/GPL 等开源协议声明
+- ❌ GitHub 仓库链接、Star History、Issue/PR 链接
+- ❌ "开源" "Open Source" 等字样
+- ❌ 贡献者列表和感谢信息 (Contributors section)
+- ❌ GitHub Badges (stars, forks, issues)
+- ❌ "Fork" "Star" "Contribute" 相关号召
+- ❌ 社区讨论链接 (Discussions, Discord, Slack)
+**保留的内容：**
+- ✅ 项目名称和描述
+- ✅ 核心功能介绍
+- ✅ 技术栈说明
+- ✅ 安装和使用文档
+- ✅ 配置说明
+- ✅ 截图和演示（如果有）
+**修改建议：**
+- 将 README.md 改为纯技术文档风格
+- 添加 "内部项目" 或 "学习项目" 标识
+- 如果有版权声明，改为 "版权所有 © [年份] 保留所有权利"
+#### 3.3 清理其他文档
+- 检查 `docs/` 目录，移除贡献指南、行为准则等开源社区文档
+- 保留技术文档、API 文档、使用手册
+### 4. Analyze Project Comprehensively
+Perform a thorough analysis of the project by examining:
+#### 4.1 Tech Stack Detection
+- Check for language-specific files:
+  - `package.json` → Node.js, inspect dependencies for React/Vue/Next.js/Express/etc.
+  - `requirements.txt`, `pyproject.toml`, `Pipfile` → Python, check for Django/Flask/FastAPI
+  - `pom.xml`, `build.gradle` → Java/Maven or Gradle
+  - `Cargo.toml` → Rust
+  - `go.mod` → Go
+  - `composer.json` → PHP
+  - `.csproj`, `*.sln` → .NET/C#
+  - `Gemfile` → Ruby
+- Identify frameworks and libraries from dependency files
+- Detect databases: check for `docker-compose.yml`, SQL files, ORM configs
+#### 4.2 Project Structure Analysis
+- List key directories and their purposes
+- Identify architecture patterns (MVC, microservices, monorepo, etc.)
+- Note important config files (`.env.example`, `config/`, `docker-compose.yml`)
+#### 4.3 Feature Extraction
+- Read modified `README.md` for feature descriptions (after cleaning open source info)
+- Look for screenshots in `docs/`, `screenshots/`, or embedded in README
+- Check for API documentation or Swagger/OpenAPI specs
+- Scan source code for major modules/features
+#### 4.4 Project Metadata
+- **注意：不要提及开源协议**，改为说明 "完整项目文件"
+- Project scale: approximate lines of code, number of files
+- Deployment info: Docker support, CI/CD configs
+### 5. Generate Three Types of Content
+**Preparation:**
+- Create a directory named `xianyu_materials` inside the project root: `<repo_name>/xianyu_materials/`
+#### 5.1 Download Project Images (提取项目演示图)
+**Step 1: Run automated download script**
+Use the provided helper script to automatically extract and download images from `README.md`.
+**Command:**
+```bash
+python .agents/skills/git-xianyu-analyzer/scripts/download_images.py <project_path>
+```
+**Actions:**
+- The script analyzes `README.md`.
+- Filters out badges and icons.
+- Downloads valid demo images to `xianyu_materials/project_images/`.
+**Step 2: Copy local project images**
+Many projects include demo images in the repository root or subdirectories. Use `find_by_name` to locate these images and copy them to `project_images/`.
+**Actions:**
+1. Search for image files: `find_by_name` with extensions `["png", "jpg", "jpeg", "gif", "webp"]` in project root (MaxDepth: 2)
+2. Exclude already processed files:
+   - Skip `xianyu_materials/` directory
+   - Skip badges/icons (e.g., small files < 10KB)
+   - Skip logo files if irrelevant to demo
+3. Copy relevant images to `xianyu_materials/project_images/`
+4. Use descriptive names like `demo_01.png`, `screenshot_dashboard.png`
+**Example:**
+```bash
+# Find images
+find . -maxdepth 2 -type f \( -name "*.png" -o -name "*.jpg" \) ! -path "./xianyu_materials/*"
+# Copy to project_images
+cp drama.png xianyu_materials/project_images/
+cp screenshot.png xianyu_materials/project_images/demo_01.png
+```
+**Note**: Some projects only have video demos (`.mp4`, `.webm`). In this case:
+- Document video links in the cleaning log
+- Users can download videos separately if needed for咸鱼 listing
+- Focus on generating AI介绍图 as the primary visual material
+#### 5.2 Deep Source Code Cleaning (LLM 源码深度清理)
+Use the Agent's capabilities to deeply scan and clean the codebase:
+1.  **Search**: Use `grep_search` to find `github.com`, `gitee.com`, `gitlab`, and other Repo URLs.
+2.  **Analyze & Clean**:
+    -   Remove `git clone` commands, repository URLs, and author links.
+    -   Replace them with generic placeholders like `https://example.com/project` or remove entirely.
+    -   **Ignore** library imports (e.g. `import github`) unless they point to the specific project's repo.
+3.  **Log**: Create a log file at `xianyu_materials/cleaning_log.md` listing every file modified and what was removed.
+#### 5.3 Compliance & Credits (合规与版权声明)
+To respect open source licenses (e.g., MIT requires preservation of copyright notice) while maintaining a "clean" product look:
+1.  **Generate NOTICE.txt**: Create a `NOTICE.txt` file in the project root.
+    -   Content: "This software is based on open source components. Copyright (c) [Year] [Original Author]. Licensed under [License Name]."
+    -   Keep this file to legally comply with permissive licenses like MIT/Apache.
+2.  **Disclaimer in Docs**: Add a small section in `docs/` or `README.md` (if kept) stating: "Core engine based on open source technology, optimized for enterprise/commercial use."
+#### 5.4 Pure Text Post (咸鱼发布用纯文本)
+Generate a concise, pure text version suitable for direct copy-paste to Xianyu:
+```
+【项目名称】[项目名]
+【技术栈】[主要技术，用空格分隔，如：Vue3 ElementPlus SpringBoot MySQL]
+【项目简介】
+[2-3 句话概括项目功能和特点]
+【核心功能】
+✅ [功能1]
+✅ [功能2]
+✅ [功能3]
+【适合场景】毕设 / 二开 / 学习
+【交付】完整源码 + 数据库 + 文档
+💰 质量保证，发货秒到！
+```
+**⚠️ 咸鱼敏感词规避规则（Critical！）:**
+**严格禁止的词汇：**
+- ❌ **源码/代码** → 改用：**项目文件/程序/系统**
+- ❌ **毕设/毕业设计/课设** → 改用：**学习项目/练手项目/参考案例**
+- ❌ **代写/定制开发** → 不要提及任何代做服务
+- ❌ **破解/激活码/授权码** → 完全避免
+- ❌ **虚拟商品/电子资料** → 改用：**技术资料/学习资料**
+- ❌ **秒到/包赔/保证/担保** → 改用：**及时发货/质量优质**
+- ❌ **加微信/QQ/私聊** → 不要留任何联系方式
+- ❌ **Git/GitHub/Gitee** → 改用：**代码仓库/项目仓库**（如需提及）
+**推荐用词替换表：**
+```
+源码 → 项目文件 / 完整项目 / 程序包
+代码 → 程序 / 文件
+毕设 → 学员作品 / 学习项目
+课设 → 练习项目 / 实战案例
+开发 → 搭建 / 编写
+数据库 → DB / 存储方案
+文档 → 说明 / 指南
+部署 → 运行 / 启动
+API → 接口
+Admin → 管理端
+```
+**文案撰写要点：**
+1. **模糊化技术性描述** - 不要过于详细的技术实现
+2. **强调学习属性** - "适合新手学习"、"技术参考"
+3. **避免商业化用词** - 不提"商用"、"企业级"
+4. **使用正向词汇** - "实战项目"、"完整案例"、"功能齐全"
+5. **不承诺服务** - 不说"包上线"、"免费维护"、"技术支持"
+**标题建议格式：**
+```
+✅ [技术栈] 实战项目 全套资料 学习专用
+✅ [项目类型] 完整程序 功能齐全
+✅ [前端框架]+[后端框架] 全栈项目 参考案例
+❌ Vue+SpringBoot源码 毕设代做 包调试
+❌ XXX管理系统 完整源码+数据库
+```
+**Save the text:**
+- Save the generated text to a file named `xianyu_materials/[项目名]_文案.txt` inside the project.
+**Requirements:**
+- Keep it concise (under 200 characters if possible)
+- 严格避免上述敏感词
+- Use euphemisms for technical terms
+- Focus on "learning" and "reference" angle
+#### 5.3 Detailed Introduction Image (详细介绍图片)
+Use the `generate_image` tool to create a professional project introduction image with:
+**Content to include:**
+- Project name (large, prominent)
+- Tech stack with icons/badges style
+- Feature list (3-5 key features)
+- Architecture diagram or feature highlights
+- Project stats (Complete Deliverable, Enterprise Arch)
+- "学习专用/二次开发/商业架构" tags
+- Visual elements: use modern gradient backgrounds, clean typography
+- **CRITICAL:** Do NOT include text like "Open Source", "MIT", "GitHub".
+**Design guidelines:**
+- Size: 1200x1600px (vertical, suitable for Xianyu)
+- Use professional color scheme (tech blues, purples, or project-relevant colors)
+- Clear hierarchy: title > tech stack > features > details
+- Include visual separators between sections
+- Use icons or badges for tech stack
+- Modern, clean aesthetic
+**Prompt template:**
+```
+A professional project introduction card for [项目名称], featuring:
+- Large bold title at top
+- Tech stack badges showing [列出技术栈]
+- Feature list with [3-5个核心功能]
+- Modern gradient background in [颜色主题]
+- Clean typography, tech industry style
+- **NO "Open Source" or "MIT" text**
+- Vertical layout 1200x1600px
+- Professional and attractive design
+```
+#### 5.4 Project Cover Image (项目封面图)
+Generate TWO outputs:
+**A. AI Prompt for Cover Image Generation:**
+Create a detailed prompt based on project type and tech stack:
+```
+Prompt for [项目名称]:
+主题：[根据项目类型，如：电商平台/管理系统/社交应用/数据分析工具等]
+视觉元素：
+- [场景描述，如：现代化的电商网站界面/专业的后台管理dashboard]
+- [技术感体现，如：代码片段/数据可视化图表/3D立体界面]
+- [色彩方案，如：科技蓝/商务紫/活力橙，根据项目定位]
+风格：
+- Modern tech illustration
+- [具体风格，如：Isometric 3D/Flat design/Glassmorphism]
+- Clean and professional
+- [项目特定元素]
+文字：无文字（或仅项目名称）
+构图：16:9 横向，适合封面展示
+```
+**B. Directly Generate Cover Image:**
+Use `generate_image` tool with the created prompt to generate the cover image.
+**Image naming:**
+- Cover image: `xianyu_materials/[项目名]_封面.png`
+**C. Save Images:**
+- Save the generated images to the `xianyu_materials/` directory created in step 5.
+### 6. Final Output
+Provide the user with:
+1. **Pure text post content** - Inside `xianyu_materials/[项目名]_文案.txt`
+3. **Project images** - Original demo images in `project_images/`
+4. **Path to generated images** - Inside `xianyu_materials` folder
+5. **AI prompt** - For reference or regeneration
+6. **Project directory path** - Where everything is saved
+Display a summary:
+```
+✅ 项目已准备完成！
+📁 项目位置: [path]
+📄 咸鱼文案（纯文本）:
+[纯文本内容]
+🖼️ 生成的图片 (位于 xianyu_materials 目录):
+- 详细介绍图: [path]
+- 项目封面图: [path]
+🎨 封面生成提示词:
+[AI prompt]
+需要调整内容或重新生成图片吗？
+```
+## Notes
+- Always verify the clone was successful before proceeding
+- Handle encoding issues when reading text files (try UTF-8, fallback to other encodings)
+- If README is very long, focus on the first few sections for feature extraction
+- Be honest about project state - don't oversell if documentation is poor

package/package-source/docs/codex-harness-engineering/templates/package-assets/skills/git-xianyu-analyzer/scripts/download_images.py ADDED Viewed

@@ -0,0 +1,112 @@
+import os
+import re
+import sys
+import time
+import urllib.request
+import urllib.parse
+from pathlib import Path
+def download_images(project_path):
+    print(f"Analyzing README in: {project_path}")
+    readme_path = Path(project_path) / "README.md"
+    if not readme_path.exists():
+        print(f"README.md not found in {project_path}")
+        return
+    try:
+        content = readme_path.read_text(encoding='utf-8', errors='ignore')
+    except Exception as e:
+        print(f"Error reading README: {e}")
+        return
+    # Regex for markdown images: ![alt](url)
+    md_images = re.findall(r'!\[.*?\]\((.*?)\)', content)
+    # Regex for HTML images: <img ... src="url" ...>
+    html_images = re.findall(r'<img[^>]+src=["\'](.*?)["\']', content)
+    all_urls = md_images + html_images
+    # Filter badges/icons
+    badge_domains = ['img.shields.io', 'travis-ci.org', 'github.com/workflows', 'badgen.net', 'api.star-history.com', 'trendshift.io']
+    valid_urls = []
+    for url in all_urls:
+        if any(b in url for b in badge_domains):
+            continue
+        if url.endswith('.svg'):
+            continue
+        valid_urls.append(url)
+    output_dir = Path(project_path) / "xianyu_materials" / "project_images"
+    output_dir.mkdir(parents=True, exist_ok=True)
+    print(f"Found {len(all_urls)} images, {len(valid_urls)} valid candidates.")
+    count = 0
+    for i, url in enumerate(valid_urls):
+        try:
+            target_url = url
+            if not target_url.startswith('http'):
+                print(f"Skipping non-http URL: {target_url}")
+                continue
+            print(f"Downloading {target_url}...")
+            # Retry logic
+            max_retries = 3
+            success = False
+            for attempt in range(max_retries):
+                try:
+                    req = urllib.request.Request(
+                        target_url,
+                        data=None,
+                        headers={
+                            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
+                        }
+                    )
+                    with urllib.request.urlopen(req, timeout=60) as response:
+                        if response.status != 200:
+                            print(f"Failed to download {target_url}: HTTP {response.status}")
+                            break # Don't retry client errors if we assume user agent is fine
+                        data = response.read()
+                        # Guess extension
+                        ext = os.path.splitext(target_url)[1].split('?')[0]
+                        if not ext:
+                            content_type = response.headers.get('Content-Type', '')
+                            if 'image/png' in content_type:
+                                ext = '.png'
+                            elif 'image/jpeg' in content_type:
+                                ext = '.jpg'
+                            else:
+                                ext = '.png'
+                        filename = f"demo_{count+1:02d}{ext}"
+                        file_path = output_dir / filename
+                        file_path.write_bytes(data)
+                        print(f"Saved: {file_path}")
+                        success = True
+                        break # Success, exit retry loop
+                except Exception as attempt_err:
+                    print(f"Attempt {attempt+1}/{max_retries} failed for {target_url}: {attempt_err}")
+                    time.sleep(2) # Wait before retry
+            if success:
+                count += 1
+            else:
+                 print(f"Permanently failed to download: {target_url}")
+        except Exception as e:
+            print(f"Error processing {url}: {e}")
+    print(f"Download complete. {count}/{len(valid_urls)} images saved to {output_dir}")
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("Usage: python download_images.py <project_path>")
+        sys.exit(1)
+    download_images(sys.argv[1])