← 返回 Skills 市场
veeicwgy

Word OCR

作者 veeicwgy · GitHub ↗ · v0.2.0 · MIT-0
cross-platform ⚠ suspicious
97
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install word-ocr
功能描述
OCR and text extraction from Word documents (.docx, .doc) using the MinerU API. This skill leverages mineru-open-api CLI to perform optical character recogni...
安全使用建议
This skill appears to do what it says (OCR Word docs via mineru-open-api) but the registry metadata is missing provenance and required-dependency declarations. Before installing, verify the mineru-open-api package on npm (check publisher, download counts, repository, and source code). Confirm the mineru.net token workflow and how the CLI expects the token (env var, config file, or interactive). Prefer installing/testing the package in an isolated environment (container or VM) rather than globally on your main system. Avoid supplying sensitive credentials until you confirm the package origin and how tokens are stored/used. If you can't verify the package/source, treat this as untrusted and do not install globally.
功能分析
Type: OpenClaw Skill Name: word-ocr Version: 0.2.0 The skill bundle provides instructions for an AI agent to perform OCR on Word documents using the 'mineru-open-api' CLI tool. It includes standard installation steps via npm and provides clear usage examples for different OCR modes. No malicious code, data exfiltration, or obfuscation was found in SKILL.md or _meta.json.
能力评估
Purpose & Capability
The skill's stated purpose (OCR Word docs via mineru-open-api) aligns with the instructions, but registry metadata lists no required binaries or credentials while the SKILL.md explicitly instructs installing an npm CLI (npm install -g mineru-open-api). The absence of a declared dependency on npm/node or the mineru CLI in the metadata is an incoherence and reduces transparency. Also the registry entry provides no homepage/source to verify the package.
Instruction Scope
The SKILL.md stays on-topic (commands only target OCR of .doc/.docx). It instructs installing a global npm package and running mineru-open-api commands that read user-supplied Word files and write output to ~/MinerU-Skill/<name>_<hash>/. There are no instructions that access unrelated system paths or exfiltrate data, but the default output path and global install are operational details users should be aware of.
Install Mechanism
There is no install spec in the registry, but the instructions require a global npm install. Installing an npm package globally executes third‑party code on the host; without a homepage/source or provenance info this is higher-risk. If the mineru-open-api package on npm is legitimate and trusted this is typical, but the lack of source links in the skill metadata is a transparency gap.
Credentials
The SKILL.md says advanced/precision OCR requires a token and links to mineru.net for token management, but the skill metadata declares no required environment variables or primary credential. It's unclear how the token is provided to the CLI (env var, config file, interactive input), which is an omission that affects whether credentials may be stored or transmitted insecurely.
Persistence & Privilege
The skill does not request persistent/always-on privileges and is user-invocable only. It doesn't instruct changes to other skills or global agent settings. No elevated platform privileges are requested.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install word-ocr
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /word-ocr 触发
  4. 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
v0.2.0
Added MinerU API integration via mineru-open-api CLI for document OCR
v0.1.0
Initial release with SEO-optimized description
元数据
Slug word-ocr
版本 0.2.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

Word OCR 是什么?

OCR and text extraction from Word documents (.docx, .doc) using the MinerU API. This skill leverages mineru-open-api CLI to perform optical character recogni... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 97 次。

如何安装 Word OCR?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install word-ocr」即可一键安装,无需额外配置。

Word OCR 是免费的吗?

是的,Word OCR 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Word OCR 支持哪些平台?

Word OCR 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 Word OCR?

由 veeicwgy(@veeicwgy)开发并维护,当前版本 v0.2.0。

💬 留言讨论