Loading tool…
Literature search agent that reads candidate papers and follows citation trails instead of ranking keywords.
Undermind is a literature search tool for researchers. Rather than returning a keyword-ranked list, it runs an extended search that reads candidate papers, judges relevance, follows citation trails and reports how thoroughly it covered the topic. It keeps workspaces and paper libraries and can alert you when new relevant work appears. It also exposes an MCP connection so Claude, ChatGPT and other agents can run searches through it. The company was founded in 2023 by two MIT physics PhDs, Tom Hartke and Joshua Ramette, and went through Y Combinator's summer 2024 batch.
Free tier covers chat, deep searches and reports at standard rate limits. Pro is $16/month billed annually and raises usage limits about tenfold; Team is $15 per person per month billed annually; Enterprise is quote-only.
Use tool ↗The search actually reads papers rather than matching keywords, which is why it surfaces work that keyword ranking buries. Runs are slow by design and the free tier's rate limits bite quickly, so it earns its keep only if literature review is a regular part of your week.
Watch out: The $16 Pro price is the annual-billing rate; paying month to month costs more.
Side-by-side fact sheets against the competitors our research names — pricing, platforms, API and both verdicts on one page.
No reviews yet — be the first to review Undermind.
Reviews are tied to your EffectHub account — one review per tool, so the rating for Undermind reflects real users.
No questions yet — ask the first one about Undermind.
Questions about Undermind are tied to your EffectHub account, so answers can reach you and the section stays free of spam.