Sheaf

Frontier document management for everyone

The wet‑ink scans. The package of 32 docs merged into one PDF. The phone photos of IDs. Drop the whole thing in.

Setup a demo workflow for me How agents connect
Sheaf / Test Environment Close

Drop your files here

or click to choose

Multiple PDF, PNG, JPEG & TIFF accepted

Selecting files opens your OS picker. Processing continues in Sheaf. No files are uploaded from this page.

Engineered for:

Cadence Bank Old National Bank Webster Bank Hanover Insurance Cincinnati Financial Stewart Title
01 / INTAKE

Accept & analyze original docs. Flag anomalies.

One PDF goes in. Sheaf finds where each document starts and ends, names it, and checks it against your expected document list. You get a manifest — 43 documents, their page spans, what is missing — before a single field is read.

2026-04-30_LYNN-ANNE-ENGLISH.PDF 412 pages · 43 documents found
  • pp. 1–6URLA — Form 1003required
  • pp. 7–9Lender loan informationrequired
  • pp. 10–12Credit report — soft pullrequired
  • pp. 13–28Bank statements — 3 accountsrequired
  • pp. 29–31Unlabelled scanreads as title commitment
  • —Certificate of insurancenot in this packet
Nothing below this line has been parsed yet

You are never billed for reading a pile you would have rejected on sight.

02 / CRITERIA

Define required docs & fields. Name what’s missing.

An expected document list is a checklist of document types, each with a schema — the fields Sheaf should pull from that type. Every document found in the packet is matched to the checklist, then read for its fields. Whatever is missing gets named: no certificate of insurance in the packet, no ending balance on a bank statement.

Expected documents Purchase loan · California

  • URLA — Form 1003signed
  • Credit reportunder 90 days
  • Bank statements3 months
  • Certificate of insurancecurrent term
  • Government IDunexpired

Schema Bank statement

  • account_holderstring
  • period_startdate
  • period_enddate
  • ending_balancecurrency
  • nsf_countinteger

Change the list and the whole corpus re‑scores against it. Nothing is read twice.

03 / GROUNDING

Fingerprint every page. Drill to the exact pixel.

Every value carries its page and a four‑number box — $9,502.93, page 3 of 6. Any number drills straight back to the pixel it was read from.

What comes back for one field

label
Borrower — monthly income
value
$9,502.93
page
3 of 6
box
0.612, 0.284, 0.788, 0.301
grounding
pdf‑text‑layer

Absence is a value. A field your schema asked for and the document did not contain comes back marked not‑stated. You can query it.

Technical detail Why a value has to point at a page, and what an agent gets when it does Read the spec →
04 / REVIEW

Human‑in‑the‑loop review: building custom intelligence.

Three decisions per document — page span, requirement match, field values — each confirmed or redirected from the keyboard. Redirects are the training signal.

The Sheaf review room. A packet is open at URLA Form 1003, pages 1 to 6 of 43 documents, with page thumbnails across the middle, a filing panel on the right showing approve, request new version and reject, and a list of six page decisions each marked inherited good. Every stage on one screen The scan itself, not a summary Approve, or send it back A decision on every page 43 documents, re‑cuttable
Page decisions read inherited good where an earlier human decision already covered those exact pages. Approving is what files the document. Each decision is bound to a page span, so re‑cutting the boundaries around it changes nothing.
TRY IT

Open a finished case. One click.

Pick a packet and it opens in the review room, in a new tab, already analyzed. Approve, reject, re‑file, ask the assistant. No form to fill in: a workspace of your own is made as you click, and this page stays where it is.

Single‑page form

Vehicle accident report

One page. Every field read, and every value pinned to the place on the page it was read from.

One PDF · 35 pages · 6 documents

A merged packet, split for you

  1. Claim noticepp. 1–3
  2. Police reportpp. 4–9
  3. Repair estimatepp. 10–15
  4. Medical recordspp. 16–27
  5. Invoicespp. 28–32
  6. Correspondencepp. 33–35

Where each document starts and ends is suggested for you. Accept a split, move it, or merge two back together.

05 / LEARNING

Correct it once. It never comes back.

Every correction is kept with who made it, what the engine proposed, and the pages it covered. Where the human departed from the machine is the signal. Two independent reviewers agreeing promotes a rule. One person clicking forty times does not.

  1. ObservedA reviewer settles something the engine was not sure about.
  2. CorroboratedA second reviewer, independently, settles it the same way.
  3. PromotedIt stops reaching the queue. Sheaf applies it and cites the two decisions that made it.
  4. RetiredA later contradiction takes it back out, and the question returns.

Learned from your reviewers 1,140 decisions

  • Prior‑year W‑2 offered against a current‑year requirementapplied 212×
  • Unlabelled scan that reads as a title commitmentapplied 341×
  • Bank statement missing its final summary pageapplied 96×
  • Any file above the $2M authority linenever automatic

What is learned stays inside your organisation and is never pooled.

06 / CORPUS

Then ask all of it at once.

Reading already happened, so a question is a lookup. What is missing, where two documents disagree, what your own average is — 3,914 files, 61 milliseconds.

Where were we invoiced above the rate we signed?

  • Meridian Freight+$4,120RC‑4471 p.1 · INV‑88213 p.2
  • Cross Valley Haulage+$3,880RC‑4402 p.1 · INV‑87994 p.1
  • Ridgeline Carriers+$2,940RC‑4519 p.2 · INV‑88407 p.3

6 discrepancies · 3,914 files read$16,765 · 61 ms

The model receives two kilobytes of facts instead of four hundred pages. Reading ten thousand files to answer the same question would be two billion tokens.

07 / ACCESS

We did not build the chat window.

Sheaf ends in the corpus. A person, an agent, or your own software asks the same question and gets the same answer, with the same pages behind it. No seat to buy.

A person

Opens the review room.

Clicks a figure, lands on the page it was read from, approves it.

An agent

Brings its own.

Claude, Codex, whatever you already run. Same corpus, same citations, no account for you to administer.

sheaf ask "missing COI on open work orders"
Your software

Makes one request.

No session to keep alive, no browser, no seat to buy.

GET /api/data/query?q=…

All three get the same answer

Yes — 3 of 208 invoices billed above their agreed rate.

  • Invoice #4417p.1$667.00+118 over
  • Invoice #4392p.2$1,240.00+305 over
  • Invoice #4390p.7$3,502.00+3,987 over

An answer, with the pages it stands on.

Point it at the pile. That is the whole configuration.

What document work actually requires

Page-level citations

Every extracted value carries the file, page and span it came from. An answer you cannot walk back to the page is an answer you cannot defend.

Scanned and photographed documents

Real work arrives as a scan of a print of a fax. A page with no text layer is the ordinary case here, not the exception that gets routed to a person.

Knowing what is missing

Completeness is the question most systems never answer. Sheaf names the requirement that nothing satisfied, which is usually what is holding the file up.

Audit trail and reviewer decisions

Every automatic outcome points back at the human decisions behind it, so an auditor can walk from a value to a page without leaving the record.

Security and data handling

Every request is checked against the role that made it, records age out on a fixed retention window, and each value stays traceable to its page.

Compare