by 安全扫描
OpenClaw
安全
high confidence这是一个仅用于指令的产品文案技能，其请求的资源与运行时指令与其声明目的相符，且不要求提供凭据、安装或系统访问权限。
评估建议
This skill appears coherent and low-risk: it only provides writing guidelines and asks for nothing from your system. Before installing, consider: (1) the skill source/homepage is listed as unknown — prefer skills from identifiable publishers if you need accountability; (2) do not submit sensitive personal or secret data (PII, credentials, proprietary text) to any skill or model unless you understand where the data will be processed or stored; (3) test the skill with non-sensitive examples to con...详细分析 ▾
✓ 用途与能力
Name and description match the SKILL.md: guidelines and steps for reviewing and rewriting UI copy. The skill declares no binaries, env vars, or config paths — which is appropriate for a text-rewriting/copywriting helper.
✓ 指令范围
SKILL.md contains detailed copywriting rules and self-checklists in Chinese; it does not instruct the agent to read system files, access external endpoints, request unrelated credentials, or exfiltrate data. All instructions are limited to producing/rewriting user-facing text.
✓ 安装机制
No install specification and no code files are present (instruction-only). No downloads, package installs, or archive extraction are requested.
✓ 凭证需求
The skill requires no environment variables, no credentials, and no config paths. There is no disproportionate request for secrets or unrelated service tokens.
✓ 持久化与权限
Flags show default behavior (always: false, user-invocable: true, model invocation allowed). The skill does not request persistent system presence or modify other skills/configs.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/3/26
v1.0.0 首次发布产品文案 Skill。 覆盖 10 个 UI 文案场景：错误/异常提示、表单/输入项、按钮/行动点、弹窗/对话框、空状态、加载/等待状态、成功反馈、权限请求、破坏性操作确认。 包含 5 条核心原则（说人话、现状+下一步、不拒绝给出路、简洁、格式规范），每条配有正反示例，附最终 5 问自检清单。
● 无害
安装命令
点击复制官方npx clawhub@latest install product-copywriting
镜像加速npx clawhub@latest install product-copywriting --registry https://cn.longxiaskill.com