ScholarFetch
Search scholarly literature, follow citation trails, fetch full text, and manage reading lists from your agent during idea and technical research.
Overview
ScholarFetch is an Idea-phase MCP server for multi-engine scholarly search, citation traversal, full text, and reading lists.
What is this MCP server?
- ScholarFetch multi-engine scholarly search, traversal, full text, and reading lists
- Remote streamable-http endpoint at laibniz-scholarfetch-web.hf.space/mcp/
- GitHub repository laibniz/scholarfetch for server version 0.2.1
- Suited to deep dives without leaving Claude Code or Cursor
- Remote MCP type streamable-http
- Published title ScholarFetch on MCP registry schema
What problem does it solve?
Founders lose days copying DOIs and PDF links between browsers when agents cannot reach structured academic search reliably.
Who is it for?
Indie builders validating technical bets, AI features, or science-adjacent products who need papers fast inside the IDE.
Skip if: Users who only need casual web search or who cannot use remote MCP endpoints on their network.
What do I get? / Deliverables
After you add the remote MCP URL, your agent can run literature workflows and return citable sources for decisions and docs.
- Curated paper results and traversal paths from multiple scholarly engines
- Reading lists and full-text extracts usable in specs and notes
Recommended MCP Servers
Journey fit
How it compares
Scholarly research MCP, not a general web browser automation or SEO content generator.
Common Questions / FAQ
Who is io.github.laibniz/scholarfetch for?
Solo builders and researchers using AI coding agents who need academic search, full text, and reading lists in one MCP server.
When should I use io.github.laibniz/scholarfetch?
Use it during Idea research or Validate scoping when you must find papers, follow citations, or build a reading list before building.
How do I add io.github.laibniz/scholarfetch to my agent?
Register the remote MCP URL https://laibniz-scholarfetch-web.hf.space/mcp/ as streamable-http in your client and reconnect.