← 返回 Skills 市场
Paper Parse
作者
Sterlingfrank1
· GitHub ↗
· v1.0.0
3439
总下载
8
收藏
18
当前安装
1
版本数
在 OpenClaw 中安装
/install paper-parse
功能描述
对用户提供的任何学术论文（PDF附件或URL）进行双模式深度研读。当用户请求分析、研读、解读或总结一篇学术论文时，使用此技能。一次性生成两份报告：Part A 面向研究者的深度专业解析，Part B 面向快速理解的核心逻辑与价值提炼。
安全使用建议
This skill appears to do what it claims: download/parse a paper, run a structured analysis, and produce two templated reports. Before installing or using it, confirm: (1) your runtime provides pdftotext (or an equivalent PDF-to-text tool) and the platform 'file' and 'message' tools the SKILL.md references, because the metadata does not list pdftotext as a required binary; (2) you are comfortable allowing the agent to download and process full PDFs (including references and figures), which may include copyrighted or sensitive content; and (3) temporary files created during analysis will reside in the agent workspace — if you need retention or deletion guarantees, verify workspace handling/policies. If pdftotext is not available, ask the skill author to declare it or provide a fallback extraction method.
功能分析
Type: OpenClaw Skill
Name: paper-parse
Version: 1.0.0
The skill is classified as suspicious due to the potential for shell injection. The `SKILL.md` file instructs the AI agent to use the `pdftotext` command-line utility to process user-provided PDF files or URLs. If the input (filename or URL) is not rigorously sanitized by the OpenClaw platform before being passed to `pdftotext`, this could allow an attacker to inject arbitrary shell commands, leading to Remote Code Execution (RCE). While `pdftotext` is a legitimate tool for the skill's stated purpose, the method of its invocation represents a significant vulnerability.
能力评估
Purpose & Capability
The skill's name, description, templates, and instructions all align: it reads PDFs/URLs, extracts text and key figures, performs analysis, and writes two templated reports. One minor inconsistency: SKILL.md explicitly instructs use of the pdftotext command and the platform 'file'/'message' tools, but the skill metadata lists no required binaries — if pdftotext is not provided by the runtime, the instruction will fail. Otherwise the requested actions (downloading PDFs, parsing, producing reports) are appropriate for the stated purpose.
Instruction Scope
Instructions are focused on paper parsing: extract full text (abstract→references), extract/inspect key figures, create a temp analysis file, then produce Part A and Part B using included templates and deliver via the message tool. The instructions do not ask for unrelated system files, secrets, or external endpoints beyond downloading a PDF from a provided URL. They do mandate thorough coverage (including references and figures), which is consistent with the skill's goal but may require large downloads and file processing.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so it does not write or execute new code on disk. That is the lowest-risk install profile.
Credentials
The skill declares no required environment variables, credentials, or config paths. Its runtime instructions reference only local template files included with the skill and temporary analysis files — proportionate to its purpose.
Persistence & Privilege
always is false and the skill does not request persistent system-wide changes or access to other skills' configurations. It creates temporary workspace files within its own context and uses the platform message tool to deliver results — standard and limited privilege for this functionality.
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 paper-parse - 安装完成后，直接呼叫该 Skill 的名称或使用
/paper-parse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 paper-parse skill:
- Deeply analyzes any academic paper (PDF or URL) and generates two types of reports in one file.
- Part A: Provides a detailed, professional, researcher-focused analysis.
- Part B: Extracts the core logic, key findings, and value in a concise, insight-driven summary.
- Ensures academic rigor, theoretical depth, and complete reproduction of the paper’s arguments and data.
- Reports always follow a dual-mode output format separated by '---' and are delivered as a single Markdown file.
- Clear workflow with mandatory full-paper read, interim structured analysis, and templated final report.
元数据
常见问题
Paper Parse 是什么？
对用户提供的任何学术论文（PDF附件或URL）进行双模式深度研读。当用户请求分析、研读、解读或总结一篇学术论文时，使用此技能。一次性生成两份报告：Part A 面向研究者的深度专业解析，Part B 面向快速理解的核心逻辑与价值提炼。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 3439 次。
如何安装 Paper Parse？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install paper-parse」即可一键安装，无需额外配置。
Paper Parse 是免费的吗？
是的，Paper Parse 完全免费（开源免费），可自由下载、安装和使用。
Paper Parse 支持哪些平台？
Paper Parse 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 Paper Parse？
由 Sterlingfrank1（@sterlingfrank1）开发并维护，当前版本 v1.0.0。
推荐 Skills