paper-harbor

Screen scholarly article metadata from ScienceDirect or CNKI into Zotero collections.

29|1|Updated Apr 30, 2026
One-click install
npx skills add https://github.com/Lucaswangzcx/paper-harbor-skill --skill paper-harbor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paper-harbor
Source: https://github.com/Lucaswangzcx/paper-harbor-skill/tree/main
Command: npx skills add https://github.com/Lucaswangzcx/paper-harbor-skill --skill paper-harbor

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires DrissionPage, and includes scripts (resource) components.

What problem does it solve?

Paper Harbor removes the manual work of searching scholarly databases, filtering papers by year and impact factor, and recording the results into Zotero for later review and citation management.

Core Features & Use Cases

  • Literature screening: Search ScienceDirect or CNKI by keywords, publication window, impact factor threshold, and target record count.
  • Metadata capture: Collect title, authors, journal, year, DOI, article URL, access notes, and screening status into structured CSV reports.
  • Zotero import: Save approved records as metadata-only journal articles into a chosen Zotero collection, one item at a time.
  • Use case: A researcher can gather a topic-specific reading list, separate strong matches from pending items, and keep a traceable output folder without downloading full text.

Quick Start

Use skill paper-harbor to search ScienceDirect or CNKI for your topic, apply your year and impact-factor rules, and save the screened metadata into the requested Zotero collection and output directory.

Frequently Asked Questions about paper-harbor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I import metadata from ScienceDirect and CNKI into Zotero?

Metadata import into Zotero requires a logged-in browser session, Zotero Desktop with Connector, and DrissionPage. The skill captures titles, authors, and DOIs from ScienceDirect or CNKI, saving them as metadata-only items into a specified Zotero collection.

Can I filter literature search results by publication year and impact factor?

Literature screening supports filtering by publication window and impact factor threshold. You can search databases by keywords, apply your year and impact-factor rules, and target a specific record count for your review workflow.

What is the best way to screen literature without downloading full text?

Screening literature without full text is handled by capturing metadata-only records. The skill collects titles, authors, journals, DOIs, and screening status into structured CSV reports, strictly avoiding full-text downloads and CAPTCHA bypasses.

Do I need DrissionPage to automate browser literature screening?

DrissionPage is a required dependency for browser automation. It drives the logged-in session to search scholarly databases, apply keyword-based screening rules, and extract metadata for serial Zotero imports.

Does the tool bypass CAPTCHAs during CNKI or ScienceDirect searches?

No, the tool explicitly avoids CAPTCHA bypasses. It requires a logged-in browser session to function properly and focuses on metadata-only imports, ensuring compliance with database access protocols.

How do I generate a traceable literature reading list with screening status?

You generate a traceable reading list by saving approved records into a specified output directory. The skill produces structured CSV reports containing article metadata, access notes, and screening status alongside the Zotero imports.