Scholar Feed
Search 600,000+ CS, AI, and ML papers with novelty signals and citations from your agent while scoping a product idea or tracking a research niche.
Overview
Scholar Feed is an MCP server for the Idea phase that surfaces 600,000+ CS, AI, and ML papers with search, novelty scores, citations, and daily watches inside your assistant.
What is this MCP server?
- Corpus of 600,000+ CS, AI, and ML papers exposed through MCP tools
- Search with novelty scores and citation context for faster triage
- Daily watch-style feeds for ongoing monitoring of topics
- Optional SF_API_KEY: anonymous 100 calls/day, keyed 1,000 calls/day per server metadata
- stdio npm scholar-feed-mcp v3.9.1 with scholarfeed.org integration
- 600,000+ CS/AI/ML papers in catalog description
- Anonymous quota 100 MCP calls/day; with SF_API_KEY 1,000 calls/day
- Server version 3.9.1
Community signal: 7 GitHub stars.
What problem does it solve?
Builders lose days manually searching arXiv and citation graphs when they need fast evidence that an AI idea is crowded or underexplored.
Who is it for?
Indie AI product founders and agent builders who want paper search and watchlists in the same session as coding and planning.
Skip if: Teams that only need non-CS market research, legal prior-art filings, or full reference managers without MCP.
What do I get? / Deliverables
After configuration, your agent can return ranked papers and watch topics so you scope features and positioning with literature-backed context.
- Curated paper lists with citations and novelty-oriented ranking
- Ongoing topic watches for CS/AI/ML areas you care about
Recommended MCP Servers
Journey fit
Idea phase is the canonical shelf because paper discovery and citation context usually happen before you commit build scope. Research subphase fits systematic literature scan, competitor technique mapping, and watching what is new in your problem domain.
How it compares
MCP research feed for CS/AI papers, not a general web search skill or automated literature-review writing pipeline.
Common Questions / FAQ
Who is Scholar Feed for?
It is for solo builders and small teams building AI or ML products who want scholarly search inside Claude Code, Cursor, or similar MCP clients.
When should I use Scholar Feed?
Use it during idea research and validate scoping when you need papers, citations, novelty signals, or daily topic watches before committing to build.
How do I add Scholar Feed to my agent?
Install scholar-feed-mcp from npm, add the stdio MCP server block in your host config, and optionally set SF_API_KEY from scholarfeed.org/settings for higher daily limits.