Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Paperclip Biomedical Literature CLI

Search and read ~11M full-text papers, FDA filings and clinical trials as a read-only virtual filesystem, with line-pinned citations.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 9/10Last updated: Aug 9, 2026

What it does

This skill teaches Claude Code to drive GXL's paperclip CLI. It exposes ~11M full-text papers (PMC, arXiv, bioRxiv, medRxiv), 217K+ FDA/PMDA/EMA regulatory documents, 110K+ trial protocols and 574K+ protein entries as a read-only virtual filesystem at /papers, /fda, /trials, /proteins.

  • Install + authentication flow (PAPERCLIP_API_KEY) with a one-call preflight check
  • Semantic search, corpus-wide grep, metadata sql, map/reduce LLM reading across many papers, and ask-image figure analysis
  • Every document is line-numbered (L1:, L2: …) so you can emit sentence-exact citation URLs like #L45
  • A rare, honest "known defects" table verified on 0.7.14/0.7.15 plus explicit warnings about commands that push data outward

Who it's for

  • Researchers writing reviews, meta-analyses or related-works sections that must cite exact lines
  • Biotech/pharma analysts comparing FDA approvals or mapping a trial landscape
  • Anyone extracting the same fields (vector, cell type, efficiency, endpoints) across dozens of papers

Examples

  1. Find then read closely: "Find 5 papers on CRISPR base-editing delivery and pull IC50 and off-target numbers with line numbers" → search -s pmc → head/scan → citation URLs.
  2. Field extraction at scale: "From 12 LNP-mRNA delivery papers, extract vector, target cell type and transfection efficiency" → search → filter → map → build the table from results.
  3. Parallel regulatory + trial sweep: "Summarise pembrolizumab accelerated approvals alongside HER2 breast cancer trials" → run -s fda and -s trials/us searches concurrently, then read meta.json.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/paperclip folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/paperclip/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/paperclip ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal on macOS or Linux (the native installer does not support Windows — use the hosted MCP server there).
  2. Clone the repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Copy the skill in: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/paperclip ~/.claude/skills/
  4. Install the CLI (this runs a remote script with your privileges — review it first): curl -fsSL https://paperclip.gxl.ai/install.sh | bash
  5. Get an API key at https://paperclip.gxl.ai/keys, save it in your project .env as PAPERCLIP_API_KEY=gxl_..., and add .env to .gitignore.
  6. Verify: run paperclip --version and paperclip config; you want to see Auth: ✓ API key (env).
  7. Restart Claude Code, then ask something like "use paperclip to find papers on ..." to trigger the skill.
  8. Be careful with upload, sync, import, share and fetch — they send local files or grant access outward, so only run them for files you explicitly name.
View source on GitHub ↗License: MIT