Connected Papers alternative
Scholar Feed vs Connected Papers: different tools, and that’s the honest answer
If you’re looking for a Connected Papers alternative, the honest take is that Scholar Feed solves a different problem. Connected Papers builds a one-shot visual graph of papers related to a seed paper — excellent for getting the lay of the land in a new area. Scholar Feed is an MCP server that searches, watches, and reads CS/AI/ML papers inside Claude Code or Cursor. If you specifically want the visual map, Connected Papers (or Litmaps) is the tool. If you want queryable, in-editor research with novelty scores and daily watches, that’s Scholar Feed. Plenty of people use both.
Why people search "Connected Papers alternative"
Connected Papers is the go-to for "show me the neighborhood of this paper" as a picture. People look for alternatives when:
- They want ongoing awareness, not a one-shot graph. Connected Papers builds a map and you’re done; it doesn’t keep telling you what’s new in that area.
- They want research in their workflow (Claude/Cursor) rather than a separate browser tool, with a relevance/novelty signal to filter the incremental flood.
How Scholar Feed compares
| Comparison axis | Connected Papers | Scholar Feed |
|---|---|---|
| Core job | Visual graph of a seed paper’s neighborhood | Search + watch + read from inside your editor |
| Shape | One-shot exploration (build a graph) | Ongoing querying and daily watches |
| Per-paper signal | Graph position | LLM summary + 0–1 novelty score |
| Keep-up | Not really its job | Daily watches on a saved filter |
| Coverage | All fields | 600,000+ CS/AI/ML papers, indexed daily |
| Form | Web app | MCP server (no UI to learn) |
What you actually do with it
You don’t get a graph from Scholar Feed. You ask your assistant "find recent high-novelty work related to 2401.04088 and summarize the top three," trace its citations both directions, pull the results section, and set a watch so new related work shows up daily. It’s the verbs around a paper (search, trace, read, monitor) as agent tools, where Connected Papers gives you the map.
When NOT to use Scholar Feed
- You want the visual graph of how papers connect. That is exactly what Connected Papers (and Litmaps) do well, and Scholar Feed does not render it. This is the common case — use Connected Papers, and reach for Scholar Feed for the search/watch/read side.
- You work outside CS/AI/ML. Connected Papers covers all fields.
- You want a zero-setup web tool. Connected Papers is click-and-go; Scholar Feed runs inside an MCP client.
Frequently asked questions
Is Scholar Feed a Connected Papers alternative?
Not a direct replacement — they solve different problems. Connected Papers builds a one-shot visual graph of a seed paper’s neighborhood. Scholar Feed is an MCP server that searches, watches, traces, and reads CS/AI/ML papers inside Claude Code or Cursor, with an LLM novelty score on each paper. Many people use both: Connected Papers for the map, Scholar Feed for ongoing querying and monitoring.
Does Scholar Feed build a visual citation graph?
No. Scholar Feed traces citations as data in both directions, but it does not render a graph you pan around. If the visual map is what you want, Connected Papers or Litmaps is the right tool.
Do I need an account or API key to try it?
No. The search and read tools work anonymously at 100 calls per day. A free API key raises the limit to 1,000 calls per day, and Pro raises it to 10,000. Install with npx scholar-feed-mcp init.
Try it
npx scholar-feed-mcp initFree anonymous access is 100 calls/day (no account); a free key raises it to 1,000/day. Open source (MIT): scholar-feed-mcp on GitHub.
More setup options on the developers page.