Academic Paper Search plugin
Paper search across five databases with deduplication, citation checks and RIS or BibTeX export for Claude Code and Codex
- 360 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The MCP server calls the CrossRef, PubMed, arXiv, OpenAlex and Europe PMC APIs and sends them your query text
- Your address from PUBMED_EMAIL is added to PubMed requests
- The CLI writes citation files and verification reports to disk
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/academic-paper-searchDetects 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 MCP server calls the CrossRef, PubMed, arXiv, OpenAlex and Europe PMC APIs and sends them your query text
- Your address from PUBMED_EMAIL is added to PubMed requests
- The CLI writes citation files and verification reports to disk
This is third-party code. Review the repository files before installing.
What it does
Academic Paper Search is a plugin for Claude Code and Codex that gathers literature on a question, verifies references and prepares files for a citation manager. It has three parts: a skill with working rules, an MCP server with four tools, and a CLI. The install name of the package stays nature-academic-search, and the word nature in it has no link to Nature Portfolio. By default the search covers CrossRef, PubMed, arXiv, OpenAlex and Europe PMC, and duplicates are merged by DOI and other strong identifiers. The search_papers tool reports which sources answered, and lookup_mesh finds MeSH terms for a PubMed query. get_citation formats one reference, and get_paper_by_id fetches a paper by DOI, PMID or arXiv ID. To check a citation, pass the expected title, author, year and journal to get_paper_by_id. Each field comes back as verified, mismatch, not_found or manual_needed. With the include_relations parameter the tool builds a graph of a paper's references and citations. The CLI writes RIS, BibTeX, NBIB and ENW files and runs YAML workflows with a verification log.
Who it is for. For graduate students and researchers who describe the task in their own words and want a verifiable source list with identifiers.
Good fit when
- You need a reading list on a topic for a review or a grant application, with a DOI or PMID for each item
- A chatbot produced references and they must be checked against real metadata
- A PubMed query should be built from MeSH terms
- You want to walk from a key paper through its reference list and the papers that cite it
- You need to export a bibliography to Zotero or EndNote as RIS, BibTeX, NBIB or ENW
Not a fit when
- You need full texts of paywalled papers: the tool works with metadata
- You need Google Scholar search: that source is not connected
- You need a guarantee that the same query a month later returns the same list: the databases keep updating
- You need the review text itself written: the plugin collects and checks sources
Example request
Use nature-academic-search to find papers on generative AI in medical education, remove duplicates and verify the top three by DOILimitations
You need Python 3.10 or newer and uv, because the plugin starts the server through uvx. For PubMed requests set a contact address in the PUBMED_EMAIL variable so that the client can read it: desktop apps do not always inherit terminal variables.
Semantic Scholar is switched on on request, and ClinicalTrials.gov works separately and serves study registrations only. The `search` and `verify` CLI commands appeared in version 0.3.1 and are missing from PyPI version 0.3.0. In workflows only the screen step calls a model automatically, and the model layer is optional: search and metadata checks work without it. Results reflect the query date, and a later rerun may return a different list.
How to disable. Remove the plugin with `claude plugin uninstall nature-academic-search@wp-a-academic-tools` in Claude Code or turn it off in Codex through the plugin manager. If you installed the CLI through uv, run `uv tool uninstall nature-academic-search`.
FAQ
How do I install Academic Paper Search in Claude Code?
Run `claude plugin marketplace add wp-a/nature-academic-search`, then `claude plugin install nature-academic-search@wp-a-academic-tools` and open a new session. You can check the network to the databases with `uvx --from nature-academic-search==0.3.1 nature-academic-search preflight`.
How do I check that a paper from a chatbot answer exists?
Call get_paper_by_id with the DOI and the expected fields. For example, DOI 10.1038/nature14539 paired with a wrong title and year resolves to the 2015 paper Deep learning, and the tool returns mismatch.
Can I search clinical trials?
Yes, the ClinicalTrials.gov source serves the trial entity type. Trial registrations are not merged with papers by title, and no bibliographic citation is built for them.
How do I get a citation graph?
Call get_paper_by_id with the include_relations parameter. The default depth is 1, and depth 2 must be requested explicitly.
Do I need a separate model key?
No, search, verification and export run without model keys. Workflows need a model only at the screen step.
Discussion
No questions or reviews yet.
Related
PluginsA Claude Code skill that turns the agent into a researcher: multi-step source collection, citation verification, and a wiki report that grows session by session
Google Scholar Skills
Google Scholar Skills for Claude Code
Six skills and a gs-researcher agent: Claude Code searches Google Scholar, tracks citations, fetches full-text links and exports to Zotero
CNKI Skills
CNKI Skills for Claude Code
A Claude Code skill set that drives CNKI through Chrome DevTools MCP: search papers, browse journals, download PDFs and export to Zotero
An MCP server that searches Google and Scholar, reads pages and PDFs and stores findings in a local project knowledge graph
Deep Research
Deep Research Skill for Claude Code
A deep research skill for Claude Code: multi-phase search, claim verification and cited reports in Markdown, HTML and PDF
Russian web search via Yandex Search API
Yandex Search
A Claude Code plugin for the official Yandex Cloud Search API: classic and generative search of the Russian web, plus a dedicated Haiku agent
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Academic Paper Search (this entry) | no | 360 | ||
| Paper Search MCP | no | key is optional | 2,950 | an MCP server for searching and downloading papers from arXiv and other databases, with no citation checks |
| arxiv-mcp-server | no | not required | 3,240 | works with arXiv only and reads papers section by section |
| Claude Scholar | no | 5,938 | a large set of skills for academic work from literature review to experiments |
README in short
The project is documented in Chinese, and its main chain is search, deduplication, verification, export. Six ready-made prompts come with worked examples, and a table shows what each database is responsible for.