← Back to Skills Marketplace
1089
Downloads
3
Stars
2
Active Installs
10
Versions
Install in OpenClaw
/install yescan-ocr-universal
Description
由夸克扫描王提供的专业 OCR 文字识别与处理工具。支持图片、截图、扫描件中的文字提取，包括手写文档、表格内容、数学公式、商品图片等复杂场景。精准识别各类证件（身份证、社保卡、驾驶证、行驶证、港澳通行证、学位证等证件）及票据（增值税发票、火车票、英文发票等票据），同时支持医疗报告单、营业执照、习题题目等专业文档识别。
Usage Guidance
This skill will read the image you provide (including local files you pass by path), encode it, and send it to Quark's API at scan-business.quark.cn for processing — so do not upload sensitive documents unless you trust the service and have reviewed its privacy/retention policy. Keep your SCAN_WEBSERVICE_KEY secret (store it in a secure vault), and restrict where you run this skill (avoid pointing it at arbitrary local paths). Be aware returned files are saved to the system temporary directory; periodically review or clean that directory. If you need stronger guarantees about data residency or retention, contact the service provider or avoid using the remote API.
Capability Analysis
Type: OpenClaw Skill
Name: yescan-ocr-universal
Version: 1.0.9
The skill bundle provides a professional OCR interface for the Quark (Alibaba) scanning service. It correctly implements input validation for URLs and local file paths in `scripts/common/validators.py`, manages API credentials securely via environment variables, and communicates with the legitimate endpoint `scan-business.quark.cn`. The code is well-structured, lacks obfuscation, and the instructions in `SKILL.md` are strictly focused on the intended OCR functionality without any signs of prompt injection or malicious intent.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description, required binary (python3), declared environment variable (SCAN_WEBSERVICE_KEY), and included client code all align with an OCR-as-a-service integration that posts images to scan-business.quark.cn. No unrelated credentials or binaries are requested.
Instruction Scope
The runtime instructions and code accept an image via URL, local file path, or base64 and will send the image data to the remote API. This matches the stated purpose, but it does mean any local file path supplied will be read, base64-encoded, and transmitted to the remote service; returned files (images/docs) may be written to the system temp directory. The SKILL.md explicitly warns about remote transmission. Users should be aware this is expected behavior and a potential data-exfiltration vector if sensitive local files are provided.
Install Mechanism
No install spec is provided (instruction-only behavior plus bundled Python scripts). Nothing is downloaded from remote URLs at install time; the code runs with python3 already on PATH. This is low-risk from an install perspective.
Credentials
Only a single credential (SCAN_WEBSERVICE_KEY) is required and is declared as the primary credential. That is proportional for a hosted OCR API. No unrelated secrets or config paths are requested.
Persistence & Privilege
always is false, the skill is user-invocable and can be called autonomously (platform default). The skill does write returned artifacts to the system temp directory but does not request persistent agent-wide privileges or modify other skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install yescan-ocr-universal - After installation, invoke the skill by name or use
/yescan-ocr-universal - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.9
- 改进了环境变量配置说明，推荐使用 OpenClaw CLI 方式进行持久配置。
- 简化和优化了主描述文本，使能力覆盖和应用场景表述更清晰。
- 技能执行命令示例从 Python subprocess 方式切换为直接 Bash 命令格式，强化约束仅允许指定结构。
- 增加了禁止自行拼接 --scene 参数的专门提示。
- 移除了 _meta.json 文件。
v1.0.8
fix: 使用 subprocess 安全参数传递，避免 shell 注入风险
v1.0.7
本地安全检查
v1.0.6
- 精简并重写了流程文档，删除全部 references/scenarios 目录及示例意图文件，改用简洁集中意图匹配规则及执行规范。
- 新增 scripts/common 目录及基础模块，优化代码结构与可维护性。
- 明确所有场景和意图的唯一匹配顺序与场景标识，并固定化执行命令格式，严禁自行拼接参数。
- 环境变量与图片输入检查流程更明确，所有异常均有唯一 JSON 响应规范。
- 统一结果直接原样透出，无需自定义包装或美化响应内容。
v1.0.5
yescan-ocr-universal 1.0.4
- No file changes detected in this version.
- Functionality, documentation, and configuration remain unchanged from the previous version.
v1.0.4
yescan-ocr-universal 1.0.4 Changelog
- No functional or content changes in this version.
- No file changes detected; this is a metadata-only release.
- All features and documented behaviors remain unchanged.
v1.0.3
v1.0.3 changelog
- No file changes detected in this release.
- Documentation, usage instructions, and skill routing logic remain unchanged.
- Version updated for release consistency; functionality and interface stay the same.
v1.0.2
yescan-ocr-universal 1.0.2 Changelog
- 更新 network 权限，加入 `https://scan-business.quark.cn/vision` 允许域名访问。
- 放宽 filesystem read 权限，允许读取“任意用户提供的本地文件路径”。
- 其余配置和调用协议、意图匹配规则等保持不变。
v1.0.1
yescan-ocr-universal 1.0.0
- 增加了重要隐私与数据流向提示，明确图片会发送至第三方服务器（夸克官方）进行 OCR 识别。
- 在「使用前必读」部分新增了安全与数据可见性说明，提升用户透明度。
- 其他内容与规范未作变更，功能和路由规则与前一版本保持一致。
v1.0.0
• 支持图片中的文字、表格、手写、证件、发票、票据、公式、营业执照、商品等多类型文字识别。
Metadata
Frequently Asked Questions
What is 夸克扫描王 OCR文字识别 - yescan ocr universal?
由夸克扫描王提供的专业 OCR 文字识别与处理工具。支持图片、截图、扫描件中的文字提取，包括手写文档、表格内容、数学公式、商品图片等复杂场景。精准识别各类证件（身份证、社保卡、驾驶证、行驶证、港澳通行证、学位证等证件）及票据（增值税发票、火车票、英文发票等票据），同时支持医疗报告单、营业执照、习题题目等专业文档识别。 It is an AI Agent Skill for Claude Code / OpenClaw, with 1089 downloads so far.
How do I install 夸克扫描王 OCR文字识别 - yescan ocr universal?
Run "/install yescan-ocr-universal" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 夸克扫描王 OCR文字识别 - yescan ocr universal free?
Yes, 夸克扫描王 OCR文字识别 - yescan ocr universal is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 夸克扫描王 OCR文字识别 - yescan ocr universal support?
夸克扫描王 OCR文字识别 - yescan ocr universal is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 夸克扫描王 OCR文字识别 - yescan ocr universal?
It is built and maintained by 夸克扫描王 (@mozhihuidage); the current version is v1.0.9.
More Skills