openalex-ajg-insights

Search ABS/AJG-ranked literature and build systematic review corpora with PDF-to-Markdown conversion.

Updated Jul 15, 2026
One-click install
npx skills add https://github.com/1549171506/shizhengmubangai1 --skill openalex-ajg-insights-1549171506
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openalex-ajg-insights
Source: https://github.com/1549171506/shizhengmubangai1/tree/main/Desktop/123/.claude/skills/openalex-ajg-insights
Command: npx skills add https://github.com/1549171506/shizhengmubangai1 --skill openalex-ajg-insights-1549171506

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, pandas, and includes scripts (resource) and references (resource) components.

What problem does it solve? Conducting a systematic literature review in business and management research requires searching ranked journals, deduplicating results, screening abstracts, downloading PDFs, and reading full texts without wasting context. This Skill automates that entire pipeline through bundled OpenAlex and paper-download backends. ## Core Features & Use Cases - ABS/AJG-Filtered Search: Query OpenAlex restricted to journals by field code and minimum rank (3, 4, 4*), with Excel and RIS exports. - Review Workspace Management: Initialize workspaces, merge raw searches into a deduplicated master corpus, maintain screening and evidence tables, and export handoff packages for Codex or Claude Code. - Full-Text Pipeline: Download priority PDFs incrementally from a manifest, convert them to Markdown via the MinerU API, chunk the Markdown, and retrieve only relevant sections by purpose (viewpoint, definition, method, finding). - Use Case: A PhD student researching entrepreneurial bricolage initializes a workspace, searches 4* journals since 2020, screens abstracts, downloads the top-cited classics, converts them with MinerU, and retrieves targeted chunks to build an evidence table. ## Quick Start Ask the AI to initialize a review workspace for your topic and search ABS/AJG-ranked journals through the bundled OpenAlex backend, then merge the results into a master corpus.

Frequently Asked Questions about openalex-ajg-insights

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

FAQPage Schema
How do I search ABS/AJG-ranked journals with OpenAlex?

Run the openalex_ajg_bridge.py script with the search-abs command, providing a query, optional field code, minimum rank (3, 4, or 4*), start year, and result limit. It returns JSON or Markdown with titles, abstracts, citations, DOIs, and full-text priority estimates.

How do I convert collected PDFs to Markdown for literature review?

Place PDFs in the workspace's 04_fulltext/pdf_inbox folder, create a mineru.env file with your MINERU_API_KEY, then run review_workflow.py with the convert-pdfs-with-mineru command. The script uploads batches to the MinerU API and stores extracted Markdown under 05_mineru/extracted.

Can I download papers automatically from the full-text manifest?

Yes, download_manifest_papers.py reads fulltext_manifest.csv and downloads PDFs by DOI through the bundled paper-download-mcp backend. It supports priority filtering, paper caps, dry runs, and incremental downloads that skip papers whose PDFs already exist.

Does the skill read entire papers when answering questions?

No, it chunks converted Markdown by headings into roughly 450-word segments and retrieves only chunks matching your query purpose, such as viewpoint, definition, method, or finding. This reduces token usage and keeps evidence extraction focused.

What credentials does the MinerU PDF conversion require?

MinerU conversion requires a MINERU_API_KEY stored in a dedicated env file such as 04_fulltext/mineru.env. Access-key and secret-key authentication is no longer supported; the token is sent as a Bearer authorization header.