paper-search-pro skill
A skill that searches five open databases for papers and builds an offline report with relevance, BibTeX and search strategies.
- 230 ★ on the repository
- Last commit
Extended access
Why this level
- The skill runs Python scripts and calls external scientific APIs with your keys.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/paper-search-proDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill runs Python scripts and calls external scientific APIs with your keys.
This is third-party code. Review the repository files before installing.
What it does
On a request like "find papers on working memory training in older adults" paper-search-pro queries OpenAlex, Semantic Scholar, CrossRef, PubMed and arXiv. A Chinese query adds NSSD and yiigle, and AI topics bring in OpenReview with ICLR, NeurIPS and ICML acceptances. Parallel subagents rate relevance and no separate LLM key is needed because the agent itself is the model. Search depth comes in four tiers. Quick gives 20–60 papers in 5–8 minutes, Standard (default) 60–180 in 10–17 minutes, Deep 180–400 with citation chasing, and Audit 400–1000 and more with a PRISMA-S log. The result is a single HTML file with three tabs that works offline, plus BibTeX for Zotero and Mendeley. For closed databases (Web of Science, Scopus, Embase, CNKI) the skill prepares search strings with the right syntax and does not query the closed databases.
Who it is for. Graduate students, researchers and review authors who need a quick literature overview on a topic from Claude Code or Codex.
Good fit when
- Before a grant proposal you need to see what is already published on the topic.
- A thesis chapter needs a BibTeX list for Zotero.
- A systematic review is being prepared with a PRISMA-S log.
- You need a search string for Web of Science or Scopus as a first draft.
Not a fit when
- You need a submission-ready systematic review: strings for closed databases remain a draft with review notes.
- You need direct access to paid databases: the skill does not query them.
- The agent cannot run Python 3.10 and parallel subagents.
Example request
Find papers on working memory training in older adults and build a Standard tier report.Limitations
Python 3.10 or newer and the dependencies from `scripts/requirements.txt` are required. Five free API keys take about 15 minutes to set up following `references/setup.md`. The Audit tier takes 2–3 hours and spends a lot of agent tokens. The skill itself calls search strings for closed databases a draft with review points, so check them with a librarian or the Cochrane protocol. The license is Apache-2.0, the last commit dates from November 2026.
How to disable. Delete the `paper-search-pro` folder from the skills directory (for example `~/.claude/skills/`) and the `PSP_HOME` variable from the shell, then restart the agent.
FAQ
How do I install paper-search-pro?
Clone the repository into the agent skills directory, for example `~/.claude/skills/paper-search-pro`, and run `python3 -m pip install -r scripts/requirements.txt`.
Is a separate LLM key required?
No, the agent itself rates relevance through parallel subagents. Only five free keys for the scientific databases are needed.
Which tier suits a course paper?
Standard: 60–180 papers in 10–17 minutes, usually enough for a literature background. A review article calls for Deep.
Which databases does the skill query?
OpenAlex, Semantic Scholar, CrossRef, PubMed and arXiv, plus NSSD and yiigle for Chinese queries and OpenReview for AI topics.
What form does the result take?
A single HTML file with three tabs that opens offline in a browser, plus BibTeX for Zotero and Mendeley.
Discussion
No questions or reviews yet.
Related
SkillsA skill that searches papers through Semantic Scholar, checks venue rankings and generates BibTeX for a manuscript.
A Claude Code skill for scientific literature reviews: an interview, Scopus and OpenAlex search, DOI checks and a final docx file
A library of 400+ academic skills: literature search, reviews, statistics, scientific writing and grant proposals
Searches Semantic Scholar, arXiv and Crossref, drafts an abstract-level review and formats DOI citations in GB/T 7714 and APA
The paper-spine skill takes a topic, materials and data through a scientific paper: literature, text, figures, review and Word, LaTeX and PDF files
Academic Automation Skill searches PubMed and CNKI and saves full texts to a folder when the user already has access.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| paper-search-pro (this card) | no | 230 | |
| Claude Scholar | no | 6,036 | Claude Scholar covers the research workflow more broadly, while paper-search-pro focuses on literature search with depth tiers and a PRISMA-S log. |
| PaperPRISM | no | 109 | PaperPRISM works with a single paper, while paper-search-pro finds and ranks papers on a topic. |
README in short
A literature search skill over OpenAlex, Semantic Scholar, CrossRef, PubMed and arXiv with four depth tiers and a report in one HTML file.