← 返回 Skills 市场
65
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install pku-claspider
功能描述
北京大学课程信息爬取 CLI (dean.pku.edu.cn 教务部 + elective.pku.edu.cn 选课网 + onlineroomse.pku.edu.cn 智云课堂)。当用户提及 claspider、课程爬虫、课程信息爬取、课程目录、全校开课、按院系/教师/关键词查课、合并课程数据 时使用此...
安全使用建议
This skill's README describes a local Rust CLI that uses your browser JWT (_token) and an existing elective session stored in ~/.config/info/elective/, but the package contains no code, no install instructions, and declares no required credentials — that mismatch is a red flag. Before installing or running anything: 1) ask the publisher for the actual binary/source and an install spec (or a link to a vetted release). 2) Confirm exactly how the tool accesses ~/.config/info/elective/ and whether it will read or modify those files. 3) Never paste your browser JWT or IAAA credentials into an unverified tool; prefer scoped API tokens or run the tool in an isolated VM/container. 4) If you need this functionality, prefer a variant that documents required secrets (and why), or obtain source code to review. Absence of regex scan findings is not assurance — the manifest/instructions themselves are the source of the concern.
能力评估
Purpose & Capability
SKILL.md describes a compiled CLI (crates/, src/*.rs) and runtime commands (claspider dean/elective/zhiyun/merge). However the skill bundle is instruction-only with no binaries, no install spec, and registry metadata lists no required binaries. That is inconsistent: the instructions assume a local 'claspider' tool exists but the skill does not provide or declare how to obtain it.
Instruction Scope
Runtime instructions explicitly reference reading/using an existing elective session directory (~/.config/info/elective/) and copying a browser JWT cookie (_token) for zhiyun. Those are sensitive local artifacts. The skill text instructs reuse of local session material even though the skill metadata declared no config paths; an agent following these instructions could be directed to access private credentials/config not declared up front.
Install Mechanism
No install specification or downloadable artifact is provided despite detailed references to source files and a CLI. This leaves unclear how the described tool would be installed or whether following the SKILL.md would cause the agent to fetch code from external locations (not specified). Instruction-only packaging plus expectation of a local binary is a coherence problem and a potential operational risk.
Credentials
The skill declares no required environment variables or credentials but the instructions require sensitive tokens/sessions in practice: a zhiyun JWT copied from the browser and reuse of the elective session stored at ~/.config/info/elective/. Those are effectively secrets but are not declared as required — the mismatch is disproportionate and should be explicitly documented and justified.
Persistence & Privilege
The skill states it has no own session storage, but instructs reuse of another tool's session directory (~/.config/info/elective/). Accessing or reusing other tools' credentials/config is a privileged action relative to a simple scraper and should be declared. The skill does not request persistent installation privileges (always:false), but its instructions imply reading user-local secrets which raises privilege concerns.
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 pku-claspider - 安装完成后，直接呼叫该 Skill 的名称或使用
/pku-claspider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
v1.0.0
- Initial release of claspider, a CLI tool for scraping and merging course information from PKU sources (dean, elective, Zhiyun).
- Supports fetching and exporting full-semester course catalogs (including course number, title, credits, teacher, schedule, and description).
- Merges data from 教务部 (dean.pku.edu.cn), 选课网 (elective.pku.edu.cn), and 智云课堂 (onlineroomse.pku.edu.cn) for accurate, comprehensive data.
- CLI commands for grabbing and merging data by term, department, category, and teacher; JSON export supported.
- Not for course selection/dropping; it’s focused on information scraping and merging only.
元数据
常见问题
PKU Claspider 是什么？
北京大学课程信息爬取 CLI (dean.pku.edu.cn 教务部 + elective.pku.edu.cn 选课网 + onlineroomse.pku.edu.cn 智云课堂)。当用户提及 claspider、课程爬虫、课程信息爬取、课程目录、全校开课、按院系/教师/关键词查课、合并课程数据 时使用此... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 65 次。
如何安装 PKU Claspider？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pku-claspider」即可一键安装，无需额外配置。
PKU Claspider 是免费的吗？
是的，PKU Claspider 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
PKU Claspider 支持哪些平台？
PKU Claspider 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 PKU Claspider？
由 wjsoj（@wjsoj）开发并维护，当前版本 v1.0.0。
推荐 Skills