← 返回 Skills 市场
zhihu-ops
作者
gumailianxin
· GitHub ↗
· v1.1.0
· MIT-0
139
总下载
0
收藏
1
当前安装
2
版本数
在 OpenClaw 中安装
/install zhihu-ops
功能描述
通过 openclaw browser 在知乎执行互动与创作类操作,包括赞同、收藏、评论、回复、发布文章、编辑文章、提问、进入创作中心、查看评论管理页,以及按目标 URL 在页面中定位按钮和输入框完成动作。当用户要求在知乎上"点个赞""评论一下""回复这条""发一篇文章""改下这篇内容""提一个问题""打开创作后...
安全使用建议
This skill appears coherent for automating Zhihu interactions and asks for nothing beyond using the platform browser. Before installing, consider: (1) the skill can publish or edit content on your Zhihu account — ensure you explicitly confirm any post/edit actions and test with drafts first; (2) confirm your OpenClaw environment provides the 'openclaw browser' and that you trust it with your logged-in session; (3) the skill suggests updating its references file during use — verify whether the platform allows the skill to write back to its own files (if you prefer no persistence, deny that capability); (4) because it makes live actions on your account, monitor results and revoke any agent ability to act autonomously if you are uncomfortable. No external credentials or downloads were requested.
功能分析
Type: OpenClaw Skill
Name: zhihu-ops
Version: 1.1.0
The zhihu-ops skill bundle is a legitimate automation tool designed to help an AI agent interact with the Zhihu platform (upvoting, commenting, and publishing articles). The instructions in SKILL.md and the element mappings in references/page-map.md are consistent with standard browser automation practices, including safety checks for login status and explicit user confirmation before performing public actions like posting content.
能力评估
Purpose & Capability
Name/description (automating Zhihu actions) align with the SKILL.md: all instructions are about navigating Zhihu pages, locating buttons/inputs, and performing interactive or publishing actions. No unrelated services, binaries, or credentials are requested.
Instruction Scope
Instructions remain focused on Zhihu interactions (like like/favorite/comment/publish/edit/ask). They explicitly require login and say not to bypass it, and they require explicit user intent before publishing/edits. One notable instruction asks the agent to persist discovered page structure back into the skill's references file ('优先把稳定规律补充回这个技能或其 references 文件'), which implies writing/updating skill metadata — this is reasonable for improving reliability but is a persistence behavior that operators should be aware of.
Install Mechanism
No install spec and no code files beyond documentation — lowest-risk model. The SKILL.md expects use of the platform 'openclaw browser' but does not try to download or install anything itself.
Credentials
The skill declares no environment variables, no credentials, and does not request unrelated secrets. All page URLs referenced are Zhihu domains and are appropriate for the stated purpose.
Persistence & Privilege
always:false and agent invocation allowed (normal). The only persistence-like instruction is to update the references file with newly discovered selectors/refs — this requires the runtime to permit writing back to the skill's files or metadata. That behavior is explainable for improving future runs, but you should confirm whether the platform allows or restricts such modifications.
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 zhihu-ops - 安装完成后，直接呼叫该 Skill 的名称或使用
/zhihu-ops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
添加知乎文章格式适配：禁止Markdown语法，支持正文小标题、工具栏加粗/列表/引用
v1.0.0
zhihu-ops v1.0.0
- 首次发布，基于 openclaw browser 自动化知乎“互动”与“创作”操作流程
- 支持点赞、收藏、评论、回复、发布/编辑文章、提问、进入创作中心等常见功能
- 优先复用已知入口、页面布局和按钮语义，降低页面解析和误操作风险
- 明确处理登录与安全边界，涉及发布/评论需确认内容再执行
- 故障处理涵盖页面结构变化、弹窗遮挡等常见阻碍
- 操作完成后标准化简明汇报执行结果
元数据
常见问题
zhihu-ops 是什么？
通过 openclaw browser 在知乎执行互动与创作类操作,包括赞同、收藏、评论、回复、发布文章、编辑文章、提问、进入创作中心、查看评论管理页,以及按目标 URL 在页面中定位按钮和输入框完成动作。当用户要求在知乎上"点个赞""评论一下""回复这条""发一篇文章""改下这篇内容""提一个问题""打开创作后... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 139 次。
如何安装 zhihu-ops？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install zhihu-ops」即可一键安装，无需额外配置。
zhihu-ops 是免费的吗？
是的，zhihu-ops 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
zhihu-ops 支持哪些平台？
zhihu-ops 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 zhihu-ops？
由 gumailianxin（@gumailianxin）开发并维护，当前版本 v1.1.0。
推荐 Skills