Hi Mrinal ,
While reviewing CyberShield, I had an idea that could improve both usability and user safety.
Currently, users need to manually copy and paste URLs for analysis. Instead, we could introduce a Screenshot Analysis feature where users simply drag and drop a screenshot of a suspicious message, email, SMS, or website.
The workflow would be:
• Upload screenshot
• OCR extracts text from the image
• URLs/domains are detected automatically
• Extracted links are sent to the phishing detection engine
• A detailed safety report is generated
This would help users avoid accidentally opening suspicious links while copying them manually and would make the platform much more user-friendly.
Additionally, we could consider a Website Preview Sandbox feature. Before a user visits a suspicious website, CyberShield could generate a safe preview screenshot of the page and display it within the platform. This would allow users to inspect the website without directly opening it, adding another layer of protection.
I believe these features would make CyberShield feel more professional, practical, and security-focused while solving real-world user problems. I'd be interested in contributing to the implementation and helping develop these improvements. I am a GSSoC 2026 contributor and would like to work on this. I'm quite comfortable with this project. Please assign it to me if it's available!
Looking forward to your thoughts.
Hi Mrinal ,
While reviewing CyberShield, I had an idea that could improve both usability and user safety.
Currently, users need to manually copy and paste URLs for analysis. Instead, we could introduce a Screenshot Analysis feature where users simply drag and drop a screenshot of a suspicious message, email, SMS, or website.
The workflow would be:
• Upload screenshot
• OCR extracts text from the image
• URLs/domains are detected automatically
• Extracted links are sent to the phishing detection engine
• A detailed safety report is generated
This would help users avoid accidentally opening suspicious links while copying them manually and would make the platform much more user-friendly.
Additionally, we could consider a Website Preview Sandbox feature. Before a user visits a suspicious website, CyberShield could generate a safe preview screenshot of the page and display it within the platform. This would allow users to inspect the website without directly opening it, adding another layer of protection.
I believe these features would make CyberShield feel more professional, practical, and security-focused while solving real-world user problems. I'd be interested in contributing to the implementation and helping develop these improvements. I am a GSSoC 2026 contributor and would like to work on this. I'm quite comfortable with this project. Please assign it to me if it's available!
Looking forward to your thoughts.