Your assistant, working from your library and answerable to it.

Ask a general assistant about your field and it answers from memory, sometimes with a plausible paper and a plausible DOI that do not exist. Add Folio to Claude, ChatGPT or any MCP client, sign in once, and it searches what you actually saved, cites what Folio formatted, verifies what it claims, and never writes into your draft.

Start free
Folio’s reference-check card rendered inside an AI assistant: a verdict for each pasted reference, the matched record, Open DOI and Save to Folio.
The reference-check card an assistant renders through the Folio connector: a verdict for each pasted reference with the matched record, so what an AI cites can be checked against the indexes.
Works in Claude ChatGPT Claude Code Cursor VS Code
Claude· Folio connected

What does my library say about retrieval practice and long-term retention?

search_librarySearching your Folio library…
Folio's search results card inside Claude: four saved sources with verification and retraction badges, a Cite button, and Open DOI.
Four saved sources speak to this. The strongest is the retrieval-practice study in Science, which found that repeated testing beat repeated studying for retention a week later(Karpicke & Roediger, 2008). One of the four is retracted, so it is flagged rather than cited.

A real search through the connector. The card is what Claude renders; the citation is what get_citation returned.

Thirty-two tools, one account

Search the library by meaning. Citations and bibliographies in APA 7, MLA 9, Chicago 17, IEEE, Harvard or Vancouver, or as BibTeX and RIS. Full source details, notes, highlights and the passages behind a claim. Drafts, readable and searchable. Radar, Discovery Agents and Surveyor results. Twenty-five tools read; seven write, and none deletes. Read the setup guide →

Keep an AI honest

check_references verifies a pasted reference list against CrossRef, OpenAlex and OpenLibrary and shows the verdicts on a card: verified, retracted, not found, or could not be checked, never “fabricated” on the strength of a missing index record. verify_quote checks a quotation against the source’s stored full text, find_passages returns the passages to cite to, list_highlights reads what you marked in the PDF, and check_claims reports the uncited claims in a draft with the library sources that could support them. None of them touches your text.

Cards inside the conversation

In Claude, a search renders as a card with every source’s verification and retraction badges, Cite for the reference on the spot, Open DOI, and Save to Folio for anything found beyond the library. A reference check renders as a second card, one row per entry with its verdict and the record it matched. Both follow Claude’s light and dark themes; clients without app support get the same results as text.

Verify before you cite

verify_reference resolves a DOI at CrossRef or searches OpenAlex by title and reports the authoritative record plus retraction status, so a reference the assistant did not take from your library gets checked instead of trusted. search_academic looks beyond your library across six scholarly indexes and marks what you already have. Every result carries its retraction status through to the assistant, so a retracted source is flagged in the answer instead of repeated back to you.

Follow the field, read the record

radar_feed pulls what is new across your Radar subscriptions, live from the indexes and deduplicated, with in-library flags. list_agents and list_agent_finds read what your Discovery Agents have kept, unseen first. authorship_summary reads a draft’s process record from its version history, get_review_report reads its published Reviewer Report, and survey_results summarises a Surveyor project per question, without identities or raw answers.

Save and organise, with permission

With the save-and-organise permission, save_source adds a paper resolved from CrossRef, retraction-checked and deduplicated against what you have; tag_source, add_to_collection and set_reading_status file it; add_source_note records a note labelled as the assistant’s; and two tools tidy agent finds. Nothing deletes, your own notes are never rewritten, and no tool writes into a document.

Sign in, no keys to paste

Folio is its own OAuth 2.1 authorization server for the connector. An assistant sends you to Folio, you see exactly what it asks for and untick what you do not want, and it receives a short-lived token you can revoke at any time. Any MCP client that follows the current specification connects the same way.

Yours only, and never your author

A connection resolves to exactly one account; another researcher’s source id is indistinguishable from one that does not exist. Nothing is connected until you approve it, Account → Connect an AI assistant shows every app that is, and there is no tool that puts text into a draft, because the writing is yours.

How it works

01

Add Folio to your assistant

Paste https://www.usefolio.co/api/mcp into Claude (Settings → Connectors) or ChatGPT (Developer mode), run one command in Claude Code, or click Add to Cursor / VS Code in Folio.

02

Sign in and choose what it may do

Folio shows you what the app is asking for. Searching and citing your library is always included; saving and organising sources, reading drafts and reading survey results are yours to allow or untick.

03

Ask about your field

Your assistant searches your real library instead of guessing, verifies references before it presents them, quotes citations Folio formatted for it, and can be asked to check its own reference list before you rely on it.

What you get

Search your library by meaning; full source details, notes and PDF highlights
Citations and bibliographies in APA 7, MLA 9, Chicago 17, IEEE, Harvard, Vancouver, BibTeX and RIS
find_passages and verify_quote: the passages behind a claim, and quotations checked against stored full text
Your drafts, readable and searchable; check_claims for the uncited claims, matched to your library
authorship_summary and get_review_report: a draft’s process record and its published Reviewer Report
verify_reference for one citation; check_references for a whole pasted list, with a results card
search_academic across OpenAlex, CrossRef, arXiv, Europe PMC, DOAJ and Semantic Scholar
radar_feed: what is new across your Radar subscriptions, with in-library flags
list_agents and list_agent_finds: what your Discovery Agents have kept, unseen first
list_surveys and survey_results: Surveyor projects summarised per question, no identities, no raw answers
save_source, tags, collections, reading status and attributed notes, only with the save-and-organise permission
No tool that deletes, and no tool that writes into a draft
OAuth 2.1 sign-in with PKCE; four permissions you can trim on the consent page
A Connected apps list with one-click disconnect, plus personal keys for CLIs
Two interactive cards in Claude: search results, and a reference check with verdicts and Save to Folio
Works as a custom connector in Claude and ChatGPT today; one command in Claude Code; one click in Cursor and VS Code

Questions, answered

How do I connect it?

In Claude: Settings → Connectors → Add custom connector, paste https://www.usefolio.co/api/mcp, click Connect and sign in to Folio when asked. In ChatGPT: Settings → Apps & Connectors → Advanced → Developer mode → Create, paste the same address, and sign in. Claude Code takes one terminal command, and Cursor and VS Code have one-click buttons in Account → Connect an AI assistant. The setup guide has every step.

What can the assistant actually do?

Read: search the library by meaning; get citations, bibliographies and full source details; read notes, highlights and the passages that support a claim; verify quotations; read and search drafts and find their uncited claims; read a draft’s authorship record and Reviewer Report; follow Radar and Discovery Agent finds; search the public indexes; verify references one at a time or as a whole list; and read Surveyor results as per-question summaries. Write, only with permission: save sources, tag them, file them in collections, set reading status, add a note marked as its own, and tidy agent finds. It can never delete anything.

Can it write my draft for me?

No, by design. An assistant connected to Folio helps you write and never writes for you, so there is no tool that puts text into a document. It can read a draft you allow it to see, find and verify sources for it, and save them to your library. The writing stays yours.

Can it change my library?

Only with the save-and-organise permission, and only in the ways listed on the consent page: saving a source, tagging it, filing it in a collection, setting its reading status, and adding a note that is labelled as the assistant’s. Nothing it does deletes a source, and your own notes are never rewritten.

What does it see when I connect?

The consent page lists exactly what the app is asking for. Searching and citing your library is always included; saving and organising sources, reading your drafts, and reading survey results are separate permissions you can untick. Every connected app, and what each may do, is listed under Account → Connect an AI assistant, with a one-click disconnect.

Does it see my survey participants?

No. The survey tools return aggregates over complete responses and the themes already coded in Surveyor: means and spreads for scaled items, option counts for choice items, theme names for open questions. Participant identities and raw open-text answers never leave Surveyor through the connector.

Can it reach anyone else’s library?

No. A connection resolves to exactly one account, and a source id belonging to another researcher is indistinguishable from one that does not exist. There is no public or cross-library search; that is deliberate, not an oversight.

What happens if I cite something retracted?

Retraction status travels with every result, verify_reference checks CrossRef’s retraction record for any citation you ask about, and check_references does the same for a whole list, so the assistant is told when a source has been retracted rather than quietly repeating it.

Is it in the Claude or ChatGPT directories?

Not yet. It works today as a custom connector in both, over the same OAuth flow the directories require; directory listings are the next step. Any MCP client that follows the current specification connects the same way.

What does it cost, and how do I turn it off?

Nothing extra. Searches, claim checks and the Radar feed are metered under the same allowances as the matching feature inside Folio, so a free account gets what a free account gets. Disconnect an app from Account → Connect an AI assistant and it stops working immediately; the others keep working.

Your sources, in the assistant you already use.

Free, opt-in, and one sign-in away. Connect it in a minute; disconnect it in a click.

Free to start · No credit card · Export everything · No ads