paper-search — Claude Code skill: what it does, how to install
Searches academic papers via OpenAlex using keywords, DOI URLs, or work identifiers. Accesses over 250 million entries without requiring an API key.
npx skills add ykdojo/paper-search --skill paper-search -a claude-codeWhat paper-search does
The skill retrieves scholarly works directly from OpenAlex. It gathers paper titles, author lists, citation counts, abstracts, research concepts, and open-access PDF links using two local bash scripts: search.sh and paper.sh.
Keyword searches handled by search.sh support pagination, custom output limits up to 200 items, and sorting by relevance, citation count, or publication date. To inspect an individual paper, paper.sh takes a DOI URL or an OpenAlex identifier, returning metadata alongside related work identifiers for citation graph exploration.
No external API token or account is necessary because OpenAlex provides free public access. Running the commands depends entirely on bash execution inside the agent workspace.
What it helps with
- find academic studies sorted by citation counts
- retrieve open-access PDF links by DOI URL
- fetch complete abstracts and author details by OpenAlex ID
- explore research citation networks using related work identifiers
Example prompts for the agent
- Search for papers on bilingual cognitive advantages sorted by cites
- Get the abstract and PDF link for https://doi.org/10.3390/brainsci8020020
- Pull full details and related works for OpenAlex ID W2789811475
How to install paper-search
1Skills CLI (Vercel)
npx skills add ykdojo/paper-search --skill paper-search -a claude-code2Manually: put the files in place
~/.claude/skills/paper-search/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/paper-search/
1Skills CLI (Vercel)
npx skills add ykdojo/paper-search --skill paper-search -a codex2Manually: put the files in place
.agents/skills/paper-search/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ykdojo/paper-search --skill paper-search -a cursor2Manually: put the files in place
.cursor/skills/paper-search/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ykdojo/paper-search paper-search2Manually: put the files in place
.github/skills/paper-search/SKILL.md1Skills CLI (Vercel)
npx skills add ykdojo/paper-search --skill paper-search -a gemini-cli2Platform CLI
gemini skills install https://github.com/ykdojo/paper-search --path skills/paper-searchManually: put the files in place
.devin/skills/paper-search/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ykdojo/paper-search --skill paper-search -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about paper-search
How do I install paper-search in Claude Code?
Run npx skills add ykdojo/paper-search --skill paper-search -a claude-code. The file will be placed in ~/.claude/skills/paper-search/SKILL.md.
What does paper-search do?
It queries OpenAlex to find scientific literature, fetching titles, authors, citations, abstracts, and direct links to open-access PDFs.
Does paper-search require an OpenAlex API key?
No, it connects to the free OpenAlex database directly without any credentials or tokens.
Source text
SKILL.md · MIT
--- name: paper-search description: Search for academic papers by keyword, or look up a specific paper by DOI or OpenAlex ID. Powered by OpenAlex (250M+ works, free, no API key). --- Search for academic papers and get details including title, authors, citation count, DOI, abstract, and open access links. Steps: 1. Find the scripts directory: `find ~/.claude -name "search.sh" -path "*/paper-search/*" 2>/dev/null | sort -V | tail -1` - This finds the script whether installed via plugin or manual setup - The `paper.sh` script is in the same directory 2. To search for papers by keyword: ``` <scripts-dir>/search.sh "your search query" [limit] [sort] [page] ``` - `limit`: number of results per page (default: 10, max: 200) - `sort`: `relevance` (default), `cites`, or `date` - `page`: page number for pagination (default: 1) 3. To look up a specific paper by DOI or OpenAlex ID: ``` <scripts-dir>/paper.sh <DOI_URL or OpenAlex_ID> ``` - Accepts full DOI URLs like `https://doi.org/10.3390/brainsci8020020` - Or OpenAlex IDs like `W2789811475` - Returns full details: authors, abstract, concepts, open access PDF link, related works Tips: - Use `relevance` sort (default) for topical searches. Use `cites` when you want landmark papers. - Be specific with queries — "bilingual cognitive advantages executive function" beats "bilingualism brain". - Use `paper.sh` to get the full abstract when search results show "Abstract: N/A". - The `related_works` IDs from `paper.sh` can be fed back into `paper.sh` to explore the citation graph. - When the user asks for scientific backing: search broadly first, pick the most relevant/cited papers, then use `paper.sh` for full details and cite as (Author, Year, Journal).
© ykdojo · MIT · ykdojo/paper-search
