← Back to Skills Marketplace
Online Ocr Skill
by
wangzhi-collab
· GitHub ↗
· v1.0.0
· MIT-0
90
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install online-ocr-skill
Description
在线OCR图片识别技能，使用免费的OCR.space API，无需安装Tesseract
Usage Guidance
这个技能看起来与其声明的用途一致，但请注意两点：1) 所有图片数据和识别后的文本会被发送到第三方服务 OCR.space（使用 public key 'helloworld' 默认有限配额），如果包含敏感信息请勿使用公共 key 或避免上传敏感图片；2) 默认会在运行目录下创建缓存（.ocr_cache），会把识别结果写到磁盘；如需更高配额或更好隐私，请注册并在命令行/代码中提供你自己的 OCR.space API key，并在不需要时删除缓存目录。除此之外，代码中未发现把数据发送到其它隐藏端点或读取/请求与 OCR 功能无关的系统凭证。
Capability Analysis
Type: OpenClaw Skill
Name: online-ocr-skill
Version: 1.0.0
The skill provides a legitimate interface for the OCR.space API to perform optical character recognition on images. The code in online_ocr.py and ocr_cli.py is well-structured, uses standard libraries (requests, pillow), and lacks any indicators of data exfiltration, malicious execution, or prompt injection. It includes helpful features like image compression and local caching in a '.ocr_cache' directory, all of which align with its stated purpose.
Capability Assessment
Purpose & Capability
名称/描述、SKILL.md 和包含的代码一致地说明使用 OCR.space API 进行 OCR。所需二进制（python）和依赖（requests, Pillow）与功能相符，没有要求与目的不相关的权限或凭证。
Instruction Scope
运行时指令和代码将图片数据（文件、字节或URL）上传到 https://api.ocr.space/parse/image ——这正是技能的核心功能，但属于将用户数据传给第三方的行为（隐私/合规关注）。代码不会访问系统上未指定的凭证或其他配置路径. 它还会在本地写入缓存（默认 .ocr_cache），这会在磁盘上保留识别结果。
Install Mechanism
无安装脚本或远程二进制下载；这是一个 instruction-plus-source 包，要求用户通过 pip 安装 requests 和 pillow（SKILL.md 提示）。没有高风险的下载来源或可执行提取步骤。
Credentials
不要求任何环境变量或秘密。默认使用公开的免费 API key 'helloworld'（文档中已说明限制）；建议用户在需要更高配额或隐私时替换为自己的 API key。
Persistence & Privilege
技能不是强制常驻（always:false），也不会修改其它技能或系统范围配置。它可在本地创建/写入一个缓存目录（默认 .ocr_cache），这是功能性行为且与目的相符。
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install online-ocr-skill - After installation, invoke the skill by name or use
/online-ocr-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Online OCR Skill 1.0.0 – Initial Release
- Provides online OCR image recognition using the free OCR.space API; no Tesseract installation required.
- Supports multiple languages, including Chinese, English, Japanese, and more.
- Offers both Python API and command-line tool for recognizing text from image files or URLs.
- Works with common image formats (PNG, JPG, JPEG, PDF).
- Default free API key included; instructions provided for registering your own.
- Includes usage examples, batch processing, integration tips, and troubleshooting advice.
Metadata
Frequently Asked Questions
What is Online Ocr Skill?
在线OCR图片识别技能，使用免费的OCR.space API，无需安装Tesseract. It is an AI Agent Skill for Claude Code / OpenClaw, with 90 downloads so far.
How do I install Online Ocr Skill?
Run "/install online-ocr-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Online Ocr Skill free?
Yes, Online Ocr Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Online Ocr Skill support?
Online Ocr Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Online Ocr Skill?
It is built and maintained by wangzhi-collab (@wangzhi-collab); the current version is v1.0.0.
More Skills