← 返回 Skills 市场
253
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install yes-md-zh
功能描述
當任務涉及修改檔案、設定、資料庫或部署時觸發。當除錯連續失敗 2 次以上時觸發。當即將猜測或假設而沒有證據時觸發（「應該是」「可能是」「我覺得」「感覺是」）。當把問題推給用戶時觸發（「請你檢查」「建議您手動」「你可能需要」）。當改完東西沒有驗證就說完成時觸發。當下結論或判定根因時觸發。當有工具卻不用時觸發（有 W...
安全使用建议
This skill is a governance/checklist (evidence-first debugging, backups, ripple-effect checks) and is internally coherent. It is instruction-only (no install, no secrets requested). However, the instructions assume the agent can run shell commands and read system state (files, env, docker, logs) and call web search; the skill does not enumerate binaries or limit which paths will be read. Before enabling: (1) confirm whether your agent runtime is allowed to execute shell commands or access sensitive files; (2) if you want to restrict file/system access, run this skill in a sandbox or disable autonomous invocation; (3) ensure any WebSearch endpoint the agent uses is acceptable for your data; and (4) test the skill in a non-production environment so its recommended commands and checks behave as you expect.
功能分析
Type: OpenClaw Skill
Name: yes-md-zh
Version: 1.1.0
The 'yes-md-zh' skill bundle is a governance framework designed to improve AI agent performance by enforcing engineering best practices such as evidence-based debugging, mandatory backups (using `cp`), and post-change verification. It provides structured instructions in `SKILL.md` to prevent 'lazy' behaviors (e.g., guessing, pushing tasks to users) and includes safety gates for deployments. No malicious code, data exfiltration, or harmful instructions were identified.
能力评估
Purpose & Capability
The name/description match the SKILL.md content: a governance checklist that enforces evidence-first debugging, backups, impact checks, and verification. It requests no binaries, env vars, or installs — consistent with being an instruction-only guidance skill.
Instruction Scope
SKILL.md tells the agent to run system inspection and verification commands (examples: cp, curl, cat, grep, lsof, docker/docker-compose checks, env, node -v) and to use WebSearch. That is appropriate for a debugging/governance checklist, but the skill does not declare required binaries or explicitly constrain which system paths may be accessed. Users should expect the agent to attempt reading files and running shell commands when this skill is followed.
Install Mechanism
No install spec and no code files — lowest-risk model: nothing written to disk by the skill itself.
Credentials
The skill declares no required environment variables or credentials and explicitly advises not to request secrets. However, its instructions recommend inspecting the runtime environment (env, node -v, which, docker ps). That means the agent may read local environment and system state at runtime even though no env vars are declared; this is consistent with the purpose but worth noting.
Persistence & Privilege
always is false, no install or config changes are requested, and the skill does not ask to modify other skills or system-wide settings. Autonomous invocation is allowed by default, which is normal for skills and not flagged here.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install yes-md-zh - 安装完成后，直接呼叫该 Skill 的名称或使用
/yes-md-zh触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.1.0
YES.md 中文版：6 層 AI 治理架構，含機器強制執行的 Hooks
元数据
常见问题
YES.md 中文版 是什么？
當任務涉及修改檔案、設定、資料庫或部署時觸發。當除錯連續失敗 2 次以上時觸發。當即將猜測或假設而沒有證據時觸發（「應該是」「可能是」「我覺得」「感覺是」）。當把問題推給用戶時觸發（「請你檢查」「建議您手動」「你可能需要」）。當改完東西沒有驗證就說完成時觸發。當下結論或判定根因時觸發。當有工具卻不用時觸發（有 W... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 253 次。
如何安装 YES.md 中文版？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install yes-md-zh」即可一键安装，无需额外配置。
YES.md 中文版 是免费的吗？
是的，YES.md 中文版 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
YES.md 中文版 支持哪些平台？
YES.md 中文版 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 YES.md 中文版？
由 tkman（@sstklen）开发并维护，当前版本 v1.1.0。
推荐 Skills