windows-screenshot-ocr
v1.0.0Windows全屏截图（自动标记鼠标位置）+ 原生OCR文字识别。完全本地运行，无需联网，无需API Key。适用于需要截图分析屏幕内容、自动化OCR识别的场景。
⭐ 0· 262·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description claim Windows-only functionality and local OCR; the included scripts implement exactly that (mss/pyautogui for screenshots, winrt-based Windows OCR). Minor mismatch: registry metadata does not declare an OS restriction even though the skill requires Windows 10/11. The scripts also hardcode a save folder on E:\ (E:\桌面\auto_screenshot), which is unusual but explainable for the stated purpose.
Instruction Scope
SKILL.md instructs the agent to run the two local Python scripts and to install required pip packages. The scripts only read screen/mouse state, create a local folder, save images, and run local OCR — they do not reference external endpoints, other system credentials, or unrelated files. Note: windows_ocr.py contains a hardcoded image_path referencing a specific timestamped file; users must edit this to point at their screenshot. The default save path is hardcoded and may not exist on many systems.
Install Mechanism
No automated install spec is provided; dependencies are installed via standard pip lines in SKILL.md (mss, pyautogui, Pillow, winrt). This is proportional to the functionality; pip installs from PyPI are expected for Python-only tools. No downloads from arbitrary URLs or archive extraction are present.
Credentials
The skill requests no environment variables, credentials, or config paths. The code uses only local APIs and writes files to a local folder. There are no calls that would exfiltrate data or require unrelated secrets.
Persistence & Privilege
The skill does not request persistent or elevated platform privileges (always is false). It does not modify other skills or global agent settings. Its behavior is limited to local file I/O and calling Windows OCR APIs.
Assessment
This skill appears to do what it says: local screenshots with a mouse marker and Windows native OCR, and it does not contain network calls or secret requests. Before installing/running: (1) ensure you're on Windows 10/11 — it won't work on other OSes; (2) review and (if desired) change the hardcoded save_folder (E:\桌面\auto_screenshot) to a path you control; (3) edit windows_ocr.py to point image_path at the screenshot you want to process (the current default filename is an example and may not exist); (4) install the pip packages (mss, pyautogui, Pillow, winrt) from PyPI and ensure Windows OCR language packs are installed; (5) if concerned about safety, run the scripts in a controlled environment (virtual machine) and monitor network activity — although no network operations are present in the code, verifying in your environment is prudent.Like a lobster shell, security has layers — review code before you run it.
latestvk97bdh1mcwc26g34jqrc3yfm3n83xehp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.