← Back to Skills Marketplace
jessy-huang

robot-paper-post

by Jessy-Huang · GitHub ↗ · v1.3.1 · MIT-0
cross-platform ⚠ suspicious
145
Downloads
2
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install robot-paper-post
Description
深度拆解机器人/具身智能论文并生成技术推文;默认使用零额外安装的插图工作流,从 arXiv HTML 与项目页提取可直接插入文章的图片和图注
Usage Guidance
总体来说这个 skill 看起来做的正是它声称的事——从 arXiv/项目页提取图与元信息并生成带图的技术推文。但在安装或运行前请注意: - 缺失脚本与元数据不一致:SKILL.md/README 多次引用 scripts/extract_arxiv_figures.ps1,但在提供的文件清单中未见该脚本;在安装或运行前确认脚本来源并在本地审查其内容,切勿在不明来源下直接执行下载的脚本。 - 网络访问与版权:skill 会抓取 arXiv、GitHub、项目页和视频并可能 hotlink 或下载图片。确认你有权重用/发布这些图片(尤其用于公众号或商业分发时),并注意遵守作者/期刊的许可要求。 - 可选的浏览器自动化:文档把浏览器截图作为可选增强,这是合理的,但若启用该路径会需要额外依赖(Playwright/Chromium 等)。不要在未审核环境中自动安装或运行这些工具。 - 检查 homepage / upstream:clawhub.json 指向一个 GitHub 仓库 URL,但元数据顶部标注 source/homepage 为 unknown/none。若打算长期使用或信任该 skill,访问该 GitHub 仓库确认维护者与脚本实现更稳妥。 - 最小权限原则:为安全起见,在沙箱或受控环境中先运行一次完整流程,验证网络请求与文件写入位置,再在生产环境中使用。 如果你希望我更深入地检查:可以提供该仓库的 homepage URL 或把 scripts/extract_arxiv_figures.ps1 的内容贴上来,我能帮你审查脚本是否存在不合理的网络/文件操作。
Capability Analysis
Type: OpenClaw Skill Name: robot-paper-post Version: 1.3.1 The skill bundle provides a workflow for analyzing robotics papers and generating social media content, but it relies on a high-risk execution pattern. Specifically, SKILL.md and README.md instruct the agent to run a PowerShell script (scripts/extract_arxiv_figures.ps1) using the '-ExecutionPolicy Bypass' flag to download images. While this is plausibly intended for the stated purpose of extracting arXiv figures without heavy dependencies like Playwright, the use of shell execution with bypassed security policies is a risky capability. Additionally, the source code for the PowerShell script itself was not included in the provided files, preventing verification of its actual network and file system operations.
Capability Assessment
Purpose & Capability
名称与描述(把机器人/具身智能论文拆解并生成带配图的技术推文)与 SKILL.md 的工作流一致:多源检索、从 arXiv/GitHub/项目页抽 Figure、生成 Markdown。唯一不一致处是文档多次引用 scripts/extract_arxiv_figures.ps1 以实现下载,但该脚本未包含在清单文件中;clawhub.json 中的 homepage 字段存在但上方元数据标注为 none,表明元数据/发布源信息不一致。
Instruction Scope
SKILL.md 明确限定操作边界:抓取 arXiv HTML、项目页、代码仓库与视频以提取图与图注,优先使用远程 arXiv 官方图片 URL;只有在必要时才使用浏览器自动化或截图。指令未请求读取本地系统凭据、Shell 历史或其它敏感配置,范围与声明的用途相符。但运行时会执行网络抓取和(可选的)脚本下载操作——这在功能上是必要的。
Install Mechanism
该 skill 是 instruction-only、无 install spec,不会在安装时下载或执行外部二进制,符合文档中“零额外安装”的设计。没有看到任何高风险的下载 URL 或自定义安装步骤。
Credentials
requires.env / required binaries / config paths 均为空,SKILL.md 也未要求任何敏感环境变量或凭证访问(仅访问公开网页)。因此请求的环境/凭据与功能复杂度成比例。
Persistence & Privilege
flags 显示 always:false 且默认允许模型调用,但这与平台常态一致。skill 不请求修改其他技能或系统配置,也不要求长期驻留权限。
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install robot-paper-post
  3. After installation, invoke the skill by name or use /robot-paper-post
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.1
Release 1.3.1: preserve 1.3.0 history and publish the reproducible arXiv figure workflow metadata update.
v1.2.0
v1.2: 新增自动截图采集功能
v1.0.0
初始版本:包含6步工作流程、术语库、团队索引、经典论文索引
Metadata
Slug robot-paper-post
Version 1.3.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is robot-paper-post?

深度拆解机器人/具身智能论文并生成技术推文;默认使用零额外安装的插图工作流,从 arXiv HTML 与项目页提取可直接插入文章的图片和图注. It is an AI Agent Skill for Claude Code / OpenClaw, with 145 downloads so far.

How do I install robot-paper-post?

Run "/install robot-paper-post" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is robot-paper-post free?

Yes, robot-paper-post is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does robot-paper-post support?

robot-paper-post is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created robot-paper-post?

It is built and maintained by Jessy-Huang (@jessy-huang); the current version is v1.3.1.

💬 Comments