A librarian for the American record · served over MCP

Ask American history. Get answers that cite their sources.

A curated, provenance-tracked knowledge base over primary sources — testimony, treaties, hearings and declassified files — connected to your AI assistant. Every claim traces back to a verbatim passage, and when the record cannot answer, it says so instead of guessing.

97 sources · 8 topics · 1778–2014 · 1,797 automated checks

What a librarian does that a search box doesn't

A reader asks a question the record answers in more than one voice. A good librarian returns the range — and refuses to average it.

“Were enslaved people treated well by kind masters?”

The WPA collected thousands of interviews with formerly enslaved people in 1936–38. They do not agree, and the disagreement is the evidence.

Alabama · T0
“us had plenty to eat”
wpa-alabama-narratives
§ Martha Bradley
Mississippi · T0
“My marster was mean an’ cruel”
wpa-mississippi-narratives
§ speaker not identified
This volume has no validated per-informant heading, so the server returns no speaker and says why, rather than guessing a name from nearby text. Someone said this about their own life and the record no longer carries who. That is a loss, not a technicality — and inventing a name would compound it.
Arkansas · T0
“Slavery was prutty bad itself but the bad time come after the war”
wpa-arkansas-narratives-pt7
§ Cal Woods

And the answer no search box gives you: this corpus cannot tell you which was typical. The interviews were collected seventy years after emancipation, mostly by white interviewers in the Jim Crow South, from elderly Black people dependent on New Deal relief. The variance is evidence about the conditions of collection as well as about slavery. The librarian returns the range and states the limit; it does not compute an average and call it history.

The government’s own record, contradicting itself

Provenance is not a footnote. Sometimes it is the finding.

1989 edition · T1
Foreign Relations of the United States, 1952–1954, Volume X: Iran
frus-iran-1951-54-ed1 Omits the covert operation almost entirely.
2017 edition · T1
Foreign Relations of the United States, 1952–1954, Iran — Second Edition
frus-iran-1951-54-ed2 Published after decades of pressure. Documents the coup.

Both editions are held, deliberately. The 1989 volume is not a worse source to be discarded — it is evidence of what an official record looked like before disclosure. A library that silently replaced it would destroy the finding.

What this server promises — and what it can’t

This distinction is measured, not asserted. The server returns it to your assistant on every session.

Enforced here — true whoever calls

  • cite resolves or errors. Never a placeholder passage.
  • Served bytes are identical to the fetched source, re-verified by checksum.
  • A citation cannot silently belong to another source.
  • For OCR’d sources, “verbatim” is computed from three engines voting word by word — never claimed.

Depends on your assistant

  • That its quotation marks contain what cite returned.
  • That its “not covered” claims reflect check_absence.
  • That a quote’s framing doesn’t invert its sense.
  • That synthesis isn’t passed off as retrieval.

The second list is measured, not polite. Running the same 35 questions with a weaker model produced five verbatim breaks and eight false “the corpus lacks this” claims — and in four of those the assistant had called the tool, been told the material was present, and claimed the gap anyway. A server cannot inspect what its caller writes. Better tools help; they do not make it true.

What’s in it

Seven topics. Deliberately narrow and deep rather than broad and thin — each states its own gaps.

TopicSourcesWhat it holds
slavery-in-their-own-words40 All 33 published WPA narrative volumes, Douglass ×3, Jacobs, Sojourner Truth, the Fugitive Slave Act
us-covert-action-abroad22 21 FRUS volumes — Iran, Guatemala, Cuba, Chile, Congo, Vietnam, Afghanistan — and the Senate study of CIA detention
racial-terror-after-slavery16 Eleven volumes of 1871 Klan hearing testimony, Wells ×3, Du Bois, Washington
indigenous-dispossession9 Black Hawk, Apess, Eastman, Pokagon, Zitkala-Ša, Jackson — and Kappler’s treaties
labor-and-capital4 Mother Jones, Debs, Goldman, Riis
immigration-and-the-city3 Antin, Addams, Riis
womens-movement2 History of Woman Suffrage Vol I, Stanton

Stated, not hidden: seven topics is not American history. Japanese incarceration, Chinese exclusion, the borderlands, the antiwar movement and almost everything after 1976 are absent. Vantages are uneven within topics too — the covert-action record is twenty-one US government compilations and no Iranian, Guatemalan or Vietnamese source. Each topic file says so, so your assistant can.

Connect it

An MCP server. No account, no key, nothing to sign.

{
  "mcpServers": {
    "usa": {
      "type": "http",
      "url": "https://usa-mcp.net/mcp"
    }
  }
}

Then ask your assistant to call orient first — it returns the rules, the corpus statistics and the promises above. Nine tools: routing, search, verbatim citation, and two that check the assistant’s own work before it answers.