Study Vault
A privacy-first research companion for university students that captures articles, videos, and notes from the web and mobile, then lets you search and auto-cite them — all stored and processed on-device.
Undergraduate and graduate students drowning in online research who need to remember and cite sources across a semester
- Browser extension + mobile share-sheet to save research sources on the fly
- Auto-generates citations in APA, MLA, and Chicago from saved URLs and PDFs
- Semantic search across your archive ("that paper on neural scaling laws...")
- Local-only storage — research never leaves the device, so schools and roommates can't snoop
AI tools for students are proliferating but most are cloud-based with real privacy concerns around unpublished work; new efficient local LLMs enable on-device semantic search for citation work.
Citation/research tooling is a heavily covered topic (multiple comparison roundups, 15+ tools listed), confirming pain, but the 'privacy-first on-device' niche itself is unproven and student demand specifically for local-only is unclear.15 Popular Citation Tools for Academic Writing ↗Top 10 AI Tools for Citations - sourcely.net ↗
Market is saturated with established free tools (Zotero open-source, Mendeley) and a wave of AI entrants (Sourcely, Anara, Atlas, Scite, Paperpile, EndNote); the on-device angle is a differentiator but the core citation/research workflow is already well-served.Zotero vs. Mendeley vs. AI: Which Reference Manager Is Best for You? ↗Best Citation Managers for Research (2026): Zotero Plus ↗
Students are notoriously price-sensitive and Zotero (free, open-source) anchors the market at $0; premium tools like EndNote/Paperpile struggle to convert students, making monetization of a niche privacy-first variant harder.Best Citation Tools 2026 — Free & Paid Compared - gentext.ai ↗
On-device AI is a clearly growing architectural trend (Dell, Google Developer Experts, arXiv MobileRAG paper), aligned with privacy regulation and local LLM improvements; trend supports multi-year relevance.On-Device RAG for App Developers: Embeddings, Vector Search and Beyond ↗MobileRAG: A Fast, Memory-Efficient, and Energy-Efficient Method for On-Device RAG ↗
Feasible but non-trivial: MobileRAG paper explicitly notes full-scale LLMs are impractical on mobile, requiring lighter models, on-disk embeddings, and significant engineering — achievable but not a weekend build.MobileRAG: A Fast, Memory-Efficient, and Energy-Efficient Method for On-Device RAG ↗