htlin222

htlin222 / openevidence-mcp

Public

Unofficial OpenEvidence MCP server with cookies.json auth, Claude/Codex/Gemini install helpers, BibTeX citation export, and Crossref validation.

github.comhtlin222openevidence-mcp bibtex claude-code codex-cli crossref gemini-cli
10
5
69% credibility
Found Apr 10, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

An unofficial tool that connects AI assistants to OpenEvidence medical search using browser-exported session cookies to fetch articles, history, and citations.

How It Works

1
🔍 Discover the Tool

You hear about a helpful add-on that lets your AI assistant pull trusted medical evidence from OpenEvidence right into your chats.

2
📥 Get It Ready

Download the tool to your computer and run the simple setup script to prepare it.

3
🍪 Share Your Login

From your browser where you're already signed into OpenEvidence, copy your session details into a file so the tool can use your account.

4
🔗 Connect to AI Helper

Link the tool to your favorite AI assistant like Claude or similar, so it knows where to reach it.

5
✅ Test the Connection

Run a quick check to confirm everything works and your account is linked properly.

💬 Ask Medical Questions

Now chat with your AI about health topics, and it fetches detailed answers with citations, sources, and even bibliography files saved for you.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is openevidence-mcp?

openevidence-mcp is an unofficial open evidence mcp server built in TypeScript that brings OpenEvidence's medical evidence search to MCP-compatible AI clients like Claude Code, Codex CLI, and Gemini CLI. You export cookies.json from your logged-in browser session for auth—no API keys or browser automation needed—and run tools like oe_ask to query questions, oe_article_get to fetch articles, or oe_history_list to review past results. It auto-saves artifacts including BibTeX citation export and Crossref validation, making evidence workflows seamless in your local AI agent setup.

Why is it gaining traction?

Unlike official APIs or heavy browser-based scrapers, this skips Playwright entirely for lightweight cookies.json auth, with one-command install helpers for macOS, Ubuntu, and Windows plus make targets to register with claude/codex/gemini clients. BibTeX export and Crossref validation add polish for citation-heavy work, standing out among unofficial github projects like github unofficial search plugins or rust unofficial github io tools. Developers grab it for quick OpenEvidence integration without vendor lock-in.

Who should use this?

Clinicians and medical researchers prompting AI agents for evidence summaries on treatments like DLBCL or HFpEF. AI tool builders extending Cursor, Continue, or Cline with OpenEvidence access. Devs evaluating unofficial github api wrappers for personal health query pipelines.

Verdict

Solid early pick for MCP users with OpenEvidence accounts—10 stars reflect niche maturity, but comprehensive docs, smoke tests, and Apache-2.0 license make it reliable for personal installs. Credibility score of 0.699999988079071% flags upstream risks; test cookies.json auth first before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.