Scholaris

Scholaris is a research paper vault: papers, AI-generated summaries, tags, a citation graph, and PDF highlighting that syncs between your browser and your account. This extension is how papers get into it without leaving the page you're reading. WHAT IT DOES Browsing a paper's page (IEEE Xplore, ACM DL, Springer, Wiley, arXiv, or almost any other publisher) or looking at a PDF directly? Click the Scholaris icon. The extension reads the page for a title, authors, DOI, and abstract, checks it against your vault, and gets out of your way with one of a few outcomes: - Already in your vault — opens straight to the paper, no extra steps. - Not in your vault yet — shows what it found and adds it with one click. No PDF required: adding just the info (title, authors, DOI, abstract) is enough to start, and it fetches a real, free (Crossref/OpenAlex) lookup for anything missing. - Viewing the PDF itself — reads the file's own metadata for a DOI (even decompressing the PDF's internal content streams when needed, since some publishers only print the DOI as page text, not as searchable metadata), matches it against your vault, and either attaches it to a paper you already added, or adds the paper and the PDF together in one step. It works through institutional proxy logins too (EZproxy-style URL rewrites), which is where a lot of PDF detection tools quietly fail. Nothing is guessed silently. A weak or ambiguous match always shows you what it found and lets you confirm or correct it — including a manual DOI field for the rare page that genuinely has none extractable. WHAT HAPPENS AFTER YOU ADD A PAPER Once a paper's in your vault, Scholaris automatically generates a structured AI summary and tags for it in the background — from the abstract if that's all you added, upgraded to the full text once a PDF is attached. From there, the web app gives you a PDF viewer with highlighting and inline notes, a citation graph connecting related papers in your library, and a chat interface to ask questions about the paper itself. WHY THE PERMISSIONS - Host permissions (all sites): paper and PDF detection has to work on whatever site you're actually reading on — there's no fixed list of publishers to restrict this to. - Scripting: reads a page's citation metadata, and — only for a PDF you already asked to check — the file's own bytes, from inside the page itself. - webRequest: observes (never blocks or modifies) the response headers of the page you're already loading, which is how PDF detection survives an institutional proxy rewriting the URL to something with no ".pdf" in it at all. - Tabs/storage: remembers what it found for the tab you're on, so the info from a paper's landing page can be reused automatically if you open its PDF a moment later. None of this runs unless you're signed into a Scholaris account, and nothing is sent anywhere until you click Add. GETTING STARTED Create a free account at debajyotihalder.in/scholaris, generate a personal access token under Settings → API Keys, and paste it into the extension's own settings page. That's the only setup required. ABOUT Scholaris is a personal project, actively developed, free to use. If something doesn't work right on a publisher's site you use, I want to hear about it.