Nature Literature Downloader

This skill turns a user's legitimate institutional access into a repeatable process for configuring, finding, downloading, and reading academic full text. It combines a first-run library-resource configuration wizard (src/, data/, scripts/configure_school.py) with browser-based download scripts (scripts/batch_download.mjs, scripts/browser_pdf_downloader.mjs) that reuse the user's already-authenticated Chrome session.

Verified routes are examples, not defaults. Every institution should start from the user's actual library resource URL, because resource portals, CAS callbacks, EZproxy, WebVPN, IP-authenticated database pages, and database detail pages reveal the live authorization path more reliably than a school name.

Access model — read this first. For a new user, do not begin by asking for the school name or by applying a preset. First ask for the library electronic-resource link they actually use. Inspect that URL to classify the route as a resource portal, CAS/SSO login, CARSI/Shibboleth, EZproxy, WebVPN, IP-authorized database page, or publisher/database detail page. School presets are optional enrichment and fallback only.

Main workflow. First configure and save the user's real library resource entry. Let the user log in through Chrome when the route reaches institutional authentication. Reuse the saved entry plus the current browser login state for later papers. For each paper, try legitimate open-access sources first; if the article is open access, download directly. Otherwise use the library route. If the library route clearly has no permission, tell the user directly instead of treating it as a generic download failure.

Chinese literature default. When the user provides a Chinese title and no DOI/PDF URL/topic route, use the CNKI route by default. Reuse the user's current Chrome library/CNKI login state, prefer the configured discovery.cnki_url entry when present, and stop for the user if CNKI or the institution asks for login, QR, CAPTCHA, SMS/OTP, or any other verification.

Browser-state principle. Authorized downloads depend on the exact browser profile where the user is logged in. If a proxy, CDP session, or browser automation tool opens a fresh profile or a different browser with no login state, do not treat the failure as missing library permission. Switch to a control path that reuses the user's active browser session, or ask the user to authenticate in the controlled browser instance.

Format principle. PDF, HTML full text, and database-native formats such as CAJ are different deliverables. If the user asks for PDF only, require a real PDF link or %PDF response and report no_authorized_pdf_found / pdf_fetch_failed when none exists. Do not save CAJ, HTML, or a login page as if it were a PDF.

First-Run Resource Configuration

For a brand-new user, ask for a library resource URL first: