Hey Joplin community! I’m Harsh Vardhan, a B.Tech CSE student at BIT Mesra, and I’ve submitted a GSoC 2026 proposal for “Automatically Label Images in Notes Using AI” for Joplin.
I’m a full-stack developer with experience in TypeScript, React, Node.js, and AI/ML APIs. I’m also an active open-source contributor, and I’ve used Joplin daily for the past two years for notes, code snippets, and study material.
My project aims to build a plugin, joplin-plugin-ai-image-labels, that scans notes, detects embedded images, and generates meaningful alt text using AI. This would improve accessibility and make image content more usable in Joplin.
Planned features include:
• Multiple AI backends for privacy and flexibility
• Async processing with progress updates
• Dry-run preview before applying changes
• Settings for API keys, language, and batch size
• Per-note or per-image rescanning options
I’d really appreciate feedback from the community and mentors on the idea, implementation, or overall proposal.
GitHub: iamlegendharsh (Harsh Vardhan) · GitHub
Email: harshvardhan00069@gmail.com
Discord: Discord