laodao-upgrade
Skill0 starsChanged 34 days agoApache-2.0View on GitHub
Use when 用户要求升级、更新或重新安装 laodao-skills,或需要把仓库中的 skills 同步配置到 Claude Code 与 Codex。
In this repository17 other files it gives its agents.
- bilibili-researchSkill
- chatgpt-share-downloadSkill
- claude-upgradeSkill
- commit-messageSkill
- docx-to-markdownSkill
- domain-availability-checkSkill
- embedded-lintSkill
- embedded-test-sopSkill
- matt-pocock-upgradeSkill
- pdf-to-markdownSkill
- standards-initSkill
- tagSkill
- trademark-basic-searchSkill
- xlsx-to-csvSkill
- x-researchSkill
- youtube-researchSkill
- zhihu-researchSkill
The file
Files
Laodao Upgrade
Overview
使用 scripts/upgrade.py 执行可重复的安全升级。脚本负责定位源码仓库、保护本地改动、快进到 origin/main,并调用仓库的 setup.sh 同步 Claude Code 与 Codex。
Workflow
- 告知用户该操作会访问 Git 远端,并写入 Claude Code / Codex / Cursor 的用户级 skills 目录。若运行环境要求批准,直接为升级命令申请批准。
用绝对路径运行(MUST NOT 先
cd进 skill 目录——Windows 上原生进程的工作目录会锁住该目录,导致setup.sh无法替换):python3 ~/.claude/skills/laodao-upgrade/scripts/upgrade.pyCodex / Cursor 侧若装在
~/.agents/skills:python3 ~/.agents/skills/laodao-upgrade/scripts/upgrade.py解释器命令是
python时把上面的python3换成python。默认定位失败时,让用户提供源码仓库路径,随后执行:
python3 ~/.claude/skills/laodao-upgrade/scripts/upgrade.py --repo <laodao-skills-repo>- 保留并向用户展示脚本及
setup.sh的输出。成功时报告版本、是否拉到新提交,以及 skills 同步结果。
Safety Contract
- 仓库有未提交或未跟踪改动时停止。报告
git status --short结果;不要自动 stash、reset、clean、commit 或覆盖。 - 仅允许
main快进到origin/main。分支错误、detached HEAD、历史分叉或命令失败时停止并报告。 - 找不到源码仓库时停止并请求
--repo;不要自行选择位置 clone。 - 安装只调用仓库自带
setup.sh。不要绕过其受管标记和同名目录保护逻辑。 - 不删除仓库外目录;孤儿 skill 的清理由
setup.sh自己处理。
Failure Handling
| 情况 | 处理 |
| 未提交改动 | 停止,列出改动,等待用户处理 |
非 main 或历史分叉 | 停止,给出当前状态,不自动切分支或改历史 |
| 网络或权限失败 | 保留原错误;按运行环境申请权限后重试同一命令 |
| 找不到 Bash | 提示 Windows 安装 Git for Windows,其他平台确保 bash 在 PATH |
| 同名 skill 未受 laodao 管理 | 尊重 setup.sh 的 skipped 结果,不覆盖 |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

