← Back to Skills Marketplace
2476
Downloads
0
Stars
12
Active Installs
1
Versions
Install in OpenClaw
/install screenshot-ocr
Description
截图 OCR 识别工具。截图→自动识别文字→复制/保存，适合提取图片内容、表格数据、验证码。
Usage Guidance
This skill appears internally consistent and runs locally: it uses Tesseract (you must install tesseract-ocr and pip packages pytesseract/pillow) and clipboard utilities (xclip on Linux, pbcopy/pbpaste on macOS) to read images and optionally copy/save recognized text. Before installing, consider: 1) The script reads any file path you pass to it and may write output files where you tell it to — avoid running it on sensitive system files. 2) Clipboard access uses platform tools and writes a temporary file (/tmp/clipboard.png) when attempting to read the clipboard; on some systems the clipboard-handling code may be buggy. 3) There are no network calls or credential requests, so it does not exfiltrate data by design. 4) If you need Windows support, the SKILL.md/script does not provide Windows clipboard handling (you will need to verify behavior). If you want added assurance, inspect or run the script locally in a sandbox before granting it wider use.
Capability Analysis
Type: OpenClaw Skill
Name: screenshot-ocr
Version: 1.0.0
The skill bundle provides a legitimate OCR utility using Tesseract and Pytesseract. The code in scripts/ocr.py handles clipboard interactions and file processing as described, with no evidence of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
Name/description (Screenshot OCR) matches the included files and requirements. The skill only needs Python, Tesseract, and optional clipboard utilities to perform OCR; nothing requested is unrelated to OCR.
Instruction Scope
SKILL.md instructs the agent to run the included scripts/ocr.py for clipboard or file OCR and to install Tesseract/pytesseract/pillow. The runtime instructions and script operate on local images, the clipboard, and write optional output files — all within the stated purpose.
Install Mechanism
There is no install spec (instruction-only with a bundled script). No remote downloads, package installs, or archive extraction are performed by the skill itself.
Credentials
The skill requests no environment variables or credentials. The _meta.json lists python3 and tesseract as required binaries, which is appropriate for OCR. No unrelated secrets or config paths are requested.
Persistence & Privilege
always is false and the skill is user-invocable. It does not attempt to modify other skills or persist credentials; it only writes output files when explicitly asked.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install screenshot-ocr - After installation, invoke the skill by name or use
/screenshot-ocr - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of screenshot-ocr.
- Enables screenshot and image file OCR for text and tables.
- Supports Chinese, English, and numeric character recognition.
- Provides one-click copy to clipboard and save as TXT.
- Offers easy command-line usage for image and clipboard input.
- Requires Python 3, Tesseract, pytesseract, and pillow.
Metadata
Frequently Asked Questions
What is Screenshot Ocr?
截图 OCR 识别工具。截图→自动识别文字→复制/保存，适合提取图片内容、表格数据、验证码。 It is an AI Agent Skill for Claude Code / OpenClaw, with 2476 downloads so far.
How do I install Screenshot Ocr?
Run "/install screenshot-ocr" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Screenshot Ocr free?
Yes, Screenshot Ocr is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Screenshot Ocr support?
Screenshot Ocr is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Screenshot Ocr?
It is built and maintained by SxLiuYu (@sxliuyu); the current version is v1.0.0.
More Skills