Today the register holds claimed people, claimed organizations, fighter records, parcels, health providers, CHRB Stewards' Rulings, CPSC Consumer Product Recalls, FDIC BankFind Institutions, KHRC Bulletin, NHTSA vPIC Manufacturers, exclusion records, and FDA Medical Device Recalls (openFDA). The search reads that and nothing else.

An evidence layer for the public record

Every public record, with its receipt.

Pansofica Corpus Labs opens the registers of the world, joins them into one record per subject, and attaches the file, the fingerprint and the date to every fact. Machines can cite it. People can check it and correct it.

Open a record Read the contract

12 public corpora · about 28.8 million records on the door · 184 registers cataloged · as of October 3, 2026

CPSC Consumer Product RecallsReceipt on file

Claire's and Icing Stores Recalls Halloween Witch Hats Due to Burn Hazard and Violation of Federal Flammability Requirement

Recall number: 24299
Title: Claire's and Icing Stores Recalls Halloween Witch Hats Due to Burn Hazard and Violation of Federal Flammability Requirement
Recall date: 2024-07-11T00:00:00
Manufacturers: []
Made in: [{"Country":"China"}]
Importers: []
Distributors: [{"CompanyID":"","Name":"Claire's and Icing Stores, of Hoffman Estates, Illinois"}]
Sold at: [{"CompanyID":"","Name":"Claire's and Icing stores nationwide from July 2023 through December 2023 for about $15."}]

All 16 fields on the record's page

Receipt

Register: CPSC Consumer Product Recalls
Publisher: U.S. Consumer Product Safety Commission
File: pages.json · 7,128,028 bytes
Published: 2026-09-24
Read: 2026-09-29
Terms: The API provides machine readable access to publicly available recall information visible on cpsc.gov. (read 2026-09-28)
Source file: not retained

sha256 1b022dde64a0f882627398f9e405f26b44b091914a78682bd2dda6d878e04495

verify: curl -sL https://www.saferproducts.gov/RestWebServices/Recall | sha256sum

The mission, in three verbs

Open. Compile. Receipt.

Every dot is a keeper of a public register.

Commissions, regulators, exchanges, federations, standards bodies. Each publishes its own file, in its own shape, on its own clock. The accent dots are the registers held today. The quieter dots are cataloged and waiting on their terms.

Held, 21 placesCataloged, 36 placesArcs, the join

Drag the globe. Positions are the publishers' countries and states from the live catalog.

What scattering costs

The most dangerous registers are the least joined.

For every corpus we measure two things. How much harm follows when a record is missed, in proportion to how many keepers hold the subject. And how much of that gap is closed today. The empty corner is the work.

Scatter risk against correction achieved, per corpus

Correction achieved runs 0 to 100 percent, left to right. Severity when a record is missed runs S1, informational, to S5, loss of life, bottom to top.

Scatter risk against correction achieved, per corpus
Corpus	Scatter risk	Severity	Keepers	Correction achieved	Achievable
claimed people	low	1	54	0%	5%
claimed organizations	low	1	27	0%	24%
fighter records	critical	5	54	21%	5%
parcels	low	2	20	19%	20%
health providers	medium	3	54	1%	5%
CHRB Stewards' Rulings	critical	5	54	1%	5%
CPSC Consumer Product Recalls	high	4	34	1%	41%
FDIC BankFind Institutions	medium	3	27	2%	24%
KHRC Bulletin	critical	5	54	1%	5%
NHTSA vPIC Manufacturers	high	4	27	2%	24%
exclusion records	medium	3	54	3%	5%
FDA Medical Device Recalls (openFDA)	critical	5	34	1%	41%

Scatter risk = weight(severity) × (1 + fragmentation bucket). Correction achieved = loaded_share * (0.5 + 0.25*joined + 0.25*rail). Achievable = publishable_share. as of October 3, 2026

Guarded data, counted

Public is not the same as open.

A register that can only be searched one name at a time, a license that forbids reuse, a PDF scanned from a printout, a list published without a date. Each is technically public and practically closed. We count them.

184

registers cataloged, with publisher, shape and terms

public on the door today

probed: readable, terms not yet read

refused by their own terms, never stored

Open, 54Non-commercial, 6Unknown, 10Refused, 5Terms not yet read, 109

The provenance contract

A register is a description, not a program.

One page describes where the file is, what a row means and what the publisher's terms say. One engine reads every page. Every record it produces carries the same receipt, and anyone can ask for it.

# one register, one page. the engine never changes.
connector_spec: "0.1"
register:
  id: cpsc_recall
  publisher: U.S. Consumer Product Safety Commission
  terms: { url: https://www.cpsc.gov/Recalls/CPSC-Recalls-Application-Program-Interface-API-Information, read_on: "2026-09-28" }
source:
  url: https://www.saferproducts.gov/RestWebServices/Recall
  file: pages.json
subject:
  kind: cpsc_recall
  key: 10000
refresh: { detect: sha256, stale_after: "2026-10-06T07:51:23.679Z" }

$ curl -s https://api.pansofica.com/sports-data/records/cpsc_recall:10000?view=provenance
{
  "register": "CPSC Consumer Product Recalls",
  "publisher": "U.S. Consumer Product Safety Commission",
  "homeUrl": "https://www.cpsc.gov/Recalls",
  "sourceUrl": "https://www.saferproducts.gov/RestWebServices/Recall",
  "fileName": "pages.json",
  "sha256": "1b022dde64a0f882627398f9e405f26b44b091914a78682bd2dda6d878e04495",
  "bytes": 7128028,
  "publishedAt": "2026-09-24",
  "fetchedAt": "2026-09-29T07:51:23.679Z",
  "terms": {
    "text": "The API provides machine readable access to publicly available recall information visible on cpsc.gov.",
    "url": "https://www.cpsc.gov/Recalls/CPSC-Recalls-Application-Program-Interface-API-Information",
    "readOn": "2026-09-28"
  },
  "retained": false,
  "staleAfter": "2026-10-06T07:51:23.679Z",
  "supersedingPending": false,
  "verify": {
    "command": "curl -sL https://www.saferproducts.gov/RestWebServices/Recall | sha256sum",
    "method": "sha256 of the published file"
  }
}

// the same door, from ChatGPT, Claude or any agent (MCP)
{ "tool": "search", "arguments": { "query": "Claire's and Icing Stores Recalls Halloween" } }
→ hits[0].id = "cpsc_recall:10000", url = "https://www.pansofica.com/records/cpsc_recall/10000"

{ "tool": "fetch", "arguments": { "id": "cpsc_recall:10000" } }
→ the record as text, its facts kept apart by grade, and the line to cite:
"CPSC Consumer Product Recalls (U.S. Consumer Product Safety Commission), pages.json, sha256 1b022dde64a0, published 2026-09-24"

The proving ground

Combat sports is where the engine runs in production today: athletic commissions publish results sheets, and one athlete competes under several of them. It is one register family among many, and the place where the bound is held against live data.

Three registers of different things, resolved by one engine

Combat sports

A record is one entry: a person as one athletic commission lists them.

22,815

records

21,035

subjects

1.08 records per subject. Subjects as resolved in the live registry.

Combat sports
What	Count
Records	22,815
Subjects	21,035
Florida State Boxing Commission	4,873
Missouri Office of Athletics	3,407
Nevada State Athletic Commission	3,071
Texas Department of Licensing and Regulation, Combative Sports	2,836
Combat Sports Commission of Western Australia	2,099
Pennsylvania State Athletic Commission	1,876
Michigan Unarmed Combat Commission	1,603
Japan Boxing Commission	1,594
Oregon State Athletic Commission	440
New Jersey State Athletic Control Board	390
British Boxing Board of Control	315
Louisiana State Athletic Commission	141
Boxing South Africa	88
Tennessee Athletic Commission	46
North Dakota Commission of Combative Sports	20
Korea Boxing Commission	16

Scholarly authors

A record is one author signature on one paper.

72,121

records

12,275

subjects

5.88 records per subject. Subjects per the benchmark’s own labels.

Scholarly authors
What	Count
Records	72,121
Subjects	12,275
KISTI	40,383
zbMATH	15,181
AMiner	7,144
Qian	6,542
PubMed	2,871

Vessel registers

A record is one row of one published permit file.

14,241

records

9,622

subjects

1.48 records per subject. Subjects per the hull identifier, which the engine never saw.

Same four pictures, three schemas

If the engine were tuned to one vertical, these columns would not line up.

Combat sports

The proving ground, in production

A record is one entry: a person as one athletic commission lists them. A subject is one person across every commission.

Records per register

Who printed how much.

Florida State Boxing Commission4,873
Missouri Office of Athletics3,407
Nevada State Athletic Commission3,071
Texas Department of Licensing and Regulation, Combative Sports2,836
Combat Sports Commission of Western Australia2,099
Pennsylvania State Athletic Commission1,876

Records per register
Register	Records
Florida State Boxing Commission	4,873
Missouri Office of Athletics	3,407
Nevada State Athletic Commission	3,071
Texas Department of Licensing and Regulation, Combative Sports	2,836
Combat Sports Commission of Western Australia	2,099
Pennsylvania State Athletic Commission	1,876
Michigan Unarmed Combat Commission	1,603
Japan Boxing Commission	1,594
Oregon State Athletic Commission	440
New Jersey State Athletic Control Board	390
British Boxing Board of Control	315
Louisiana State Athletic Commission	141
Boxing South Africa	88
Tennessee Athletic Commission	46
North Dakota Commission of Combative Sports	20
Korea Boxing Commission	16

Records resolving into subjects

Many entries in, fewer subjects out.

22,815

records

21,035

subjects

1.08 records per subject. Subjects as resolved in the live registry.

Records resolving into subjects
What	Count
Records	22,815
Subjects	21,035
Florida State Boxing Commission	4,873
Missouri Office of Athletics	3,407
Nevada State Athletic Commission	3,071
Texas Department of Licensing and Regulation, Combative Sports	2,836
Combat Sports Commission of Western Australia	2,099
Pennsylvania State Athletic Commission	1,876
Michigan Unarmed Combat Commission	1,603
Japan Boxing Commission	1,594
Oregon State Athletic Commission	440
New Jersey State Athletic Control Board	390
British Boxing Board of Control	315
Louisiana State Athletic Commission	141
Boxing South Africa	88
Tennessee Athletic Commission	46
North Dakota Commission of Combative Sports	20
Korea Boxing Commission	16

Subjects by records carried

How many registers or rows one subject spans.

The public payload gives this split and no finer one, so none is drawn.

Subjects by records carried
Records carried	Subjects
1	19,475
2 or more	1,560

Coverage, register by facet

By discipline. Pick a column to sort and read the numbers.

Florida State Boxing Commission

Missouri Office of Athletics

Nevada State Athletic Commission

Texas Department of Licensing and Regulation, Combative Sports

Combat Sports Commission of Western Australia

Pennsylvania State Athletic Commission

fewermore

Coverage, register by facet
Register	Boxing	MMA	Kickboxing	Muay Thai	Other
Florida State Boxing Commission	2,524	1,777	62	16	255
Missouri Office of Athletics	542	1,847	1,200	270	6
Nevada State Athletic Commission	1,085	1,910	0	14	124
Texas Department of Licensing and Regulation, Combative Sports	1,181	1,445	91	187	0
Combat Sports Commission of Western Australia	338	512	0	1,089	383
Pennsylvania State Athletic Commission	703	812	381	0	0
Michigan Unarmed Combat Commission	565	839	88	195	0
Japan Boxing Commission	1,594	0	0	0	0
Oregon State Athletic Commission	0	436	20	0	0
New Jersey State Athletic Control Board	390	0	0	0	0
British Boxing Board of Control	315	0	0	0	0
Louisiana State Athletic Commission	141	0	0	0	0
Boxing South Africa	88	0	0	0	0
Tennessee Athletic Commission	46	0	0	0	0
North Dakota Commission of Combative Sports	0	20	0	0	0
Korea Boxing Commission	16	0	0	0	0

Our own public API, read when this page was built.

Scholarly authors

A public benchmark

A record is one author signature on one paper. A subject is one researcher, per the labels that ship with the benchmark.

Records per register

Who printed how much.

KISTI40,383
zbMATH15,181
AMiner7,144
Qian6,542
PubMed2,871

Records per register
Register	Records
KISTI	40,383
zbMATH	15,181
AMiner	7,144
Qian	6,542
PubMed	2,871

Records resolving into subjects

Many entries in, fewer subjects out.

72,121

records

12,275

subjects

5.88 records per subject. Subjects per the benchmark’s own labels.

Records resolving into subjects
What	Count
Records	72,121
Subjects	12,275
KISTI	40,383
zbMATH	15,181
AMiner	7,144
Qian	6,542
PubMed	2,871

Subjects by records carried

How many registers or rows one subject spans.

The largest subject carries 458 records.

Subjects by records carried
Records carried	Subjects
1	5,185
2	1,812
3 to 5	2,496
6 to 20	2,078
21 to 100	650
Over 100	54

Coverage, register by facet

By publication decade. Pick a column to sort and read the numbers.

KISTI

zbMATH

AMiner

Qian

PubMed

fewermore

Coverage, register by facet
Register	Before 1980	1980s	1990s	2000s	2010s
KISTI	385	1,588	9,987	28,403	20
zbMATH	6,852	3,550	4,453	215	111
AMiner	46	195	1,069	5,810	24
Qian	21	62	500	5,517	442
PubMed	16	33	187	1,674	961

S2AND author disambiguation benchmark, Allen Institute for AI, ODC-By 1.0.

Vessel registers

The second vertical, no engine changes

A record is one row of one published permit file. A subject is one vessel, per the hull identifier both programs print.

Records per register

Who printed how much.

Atlantic highly migratory species permits, 20266,212
Greater Atlantic vessel permits, 20254,169
Greater Atlantic vessel permits, 20263,860

Records per register
Register	Records
Atlantic highly migratory species permits, 2026	6,212
Greater Atlantic vessel permits, 2025	4,169
Greater Atlantic vessel permits, 2026	3,860

Records resolving into subjects

Many entries in, fewer subjects out.

14,241

records

9,622

subjects

1.48 records per subject. Subjects per the hull identifier, which the engine never saw.

Records resolving into subjects
What	Count
Records	14,241
Subjects	9,622
Atlantic highly migratory species permits, 2026	6,212
Greater Atlantic vessel permits, 2025	4,169
Greater Atlantic vessel permits, 2026	3,860

Subjects by records carried

How many registers or rows one subject spans.

The largest subject carries 3 records.

Subjects by records carried
Records carried	Subjects
1	5,838
2	2,949
3	835

Coverage, register by facet

By hull length. Pick a column to sort and read the numbers.

Atlantic highly migratory species permits, 2026

Greater Atlantic vessel permits, 2025

Greater Atlantic vessel permits, 2026

fewermore

Coverage, register by facet
Register	Under 25 ft	25 to 39 ft	40 to 64 ft	65 to 99 ft	100 ft and over	Not printed
Atlantic highly migratory species permits, 2026	938	3,835	1,296	137	6	0
Greater Atlantic vessel permits, 2025	380	1,657	1,473	577	37	45
Greater Atlantic vessel permits, 2026	323	1,480	1,386	564	38	69

NOAA Fisheries permit registers. Courtesy: National Oceanic and Atmospheric Administration.

Records by year, scholarly authors

Drag the handle, or use the arrow keys, to see what was on file through a given year.

30,174records on file through 2000

2,560 dated 2000. 43% of the archive. Benchmark signature files.

Year 2000

Records by year, scholarly authors
Year	Records
1950	106
1951	105
1952	129
1953	92
1954	96
1955	111
1956	104
1957	102
1958	118
1959	122
1960	139
1961	163
1962	184
1963	166
1964	177
1965	226
1966	198
1967	278
1968	274
1969	260
1970	308
1971	230
1972	227
1973	227
1974	275
1975	263
1976	294
1977	310
1978	351
1979	355
1980	388
1981	414
1982	423
1983	515
1984	495
1985	542
1986	604
1987	645
1988	648
1989	754
1990	853
1991	1,007
1992	1,057
1993	1,317
1994	1,447
1995	1,615
1996	1,842
1997	2,148
1998	2,377
1999	2,533
2000	2,560
2001	2,891
2002	3,748
2003	4,707
2004	6,434
2005	7,122
2006	6,561
2007	3,374
2008	2,219
2009	2,003
2010	763
2011	257
2012	229
2013	194
2014	62
2015	10
2016	9
2017	7
2018	7
2019	18
2020	2

From a printed line to a record row

Published document

as a register printed it

line 14, name as printed, date, outcome

Subject: one id, across every register
Fact: as printed, never edited to agree
Register: who published it
Document: address and hash of the file
Receipt: written down, reversible

From a printed line to a record row
Field	What it holds
Subject	one id, across every register
Fact	as printed, never edited to agree
Register	who published it
Document	address and hash of the file
Receipt	written down, reversible

One engine, any register

The engine does not know what a fighter, an author or a vessel is. A schema tells it what a record looks like in that register, and a mapper turns a printed row into a record.

Combat sports is where it runs in production. Scholarly authors is a public benchmark with published results to stand beside. Vessel registers ran as the second vertical with no change to the engine.

Pansophia

In the 1600s Comenius called it pansophia: all knowledge, put in order so that anyone could use it.

Our version is narrower and checkable. Take every register that lists a subject. Keep what each one printed. Join the entries that are the same subject, show the document behind each fact, and measure how often the join is wrong.

What a record carries

The same five parts in every register, whatever the subject is.

Subject: One id for one person, vessel or entity, across every register that lists it.
Facts: Each fact as the register printed it. Two registers that disagree are both shown.
Document: The published file behind every fact, by address and by hash, so you can show the sheet and not our word.
Receipt: Every join, split and correction is written down with its reason and can be reversed.
Bound: The error rate is measured on labeled data and held under a stated limit.

Measured on public data

On a public author benchmark the engine, given a schema and a mapper, is level with the published in-domain model on four of five datasets and three points behind on the fifth, under a simpler protocol.

Scholarly authors, five public datasets

B-cubed F1 on held-out test blocks, one seed, a split of our own. It is a result on public data, not a replication of the paper.

Exact-name matching
Published in-domain model
Measured on the same test blocks

Scholarly authors, five public datasets
Dataset	Exact-name matching	Published in-domain model	Measured on the same test blocks
AMiner	0.496	0.872	0.877
KISTI	0.942	0.954	0.955
PubMed	0.745	0.910	0.925
Qian	0.772	0.944	0.915
zbMATH	0.837	0.953	0.968

Vessel registers

On the vessel registers there is no published model to compare with, so the comparison is exact-name matching, checked against a hull identifier the engine never saw.

Exact-name matching
Measured on the same test blocks

Vessel registers
Dataset	Exact-name matching	Published in-domain model	Measured on the same test blocks
NOAA permit registers	0.917	none published	0.995

In the proving ground the error rate is measured against labeled rulings and held under a stated bound. The figure and how it is measured are shared under contract.

Who this is for

Three asks. One record.

For laboratories and agent builders

Ground answers about people, products and companies in records that carry a file, a fingerprint and a date, and show that receipt to the user. One door, every register, the same shape.

The door

For keepers of registers

Publish the whole file, with a date and clear terms, in a form a machine can read. A register that can only be searched one name at a time is a register the world cannot check.

The catalog

For the person on the record

If the record is about you, you can claim it, flag it or dispute it. The correction sits beside the record. The record itself is never rewritten.

Claim your record

Try the receipt

A fingerprint is what verify computes.

Type anything. Your browser computes its SHA-256, the same function the engine runs over every published file. Change one character and watch the fingerprint change entirely. That is why a receipt cannot be edited quietly.

Input

sha256 computing

Edit the text above.

Research and press

The evidence, written down.

The evidence layer

The white paper: what a register is, the invariants every record keeps, how to check us, what we ask. Published when counsel has read it.

In preparation

Press

The releases, each with its evidence note and every source.

Read the releases

Coverage

Every register we hold or have cataloged, with the file behind each dot.

See the coverage

Where it is used

The same engine, pointed at other registers. These are illustrations. None of them is a customer, and no figure is claimed for any of them.

For AI labs and teams that run agents

An agent that writes, a corpus that repeats itself, an assistant that cites, a queue of labels. Each of them is a register.

A governor for an agent that writes

The register: The system of record your agent writes to: a customer file, a ledger, a case system.
A subject: One entity in that system. Every write the agent makes to it is a record.
What the engine gives: Every write is kept with a snapshot of what the agent saw, a receipt and a way back. A new version replays past decisions before it ships, the error rate is measured against a stated bound, and a write that cannot be made inside that bound is withheld for a person.

An update dated after the entity was closed

One subject, counted once, in a corpus

The register: A training or evaluation corpus drawn from many sources.
A subject: One author, organization, product or place, under however many spellings it arrives.
What the engine gives: Near-duplicates resolve into one subject, so nothing is counted twice and an evaluation set is not contaminated by items the training set already holds. Every join has a receipt and can be reversed.

A held-out item that training had already seen

Provenance for retrieval

The register: The document store an assistant retrieves from.
A subject: Whatever the assistant is asked about. A record is one fact as one document states it.
What the engine gives: Every fact the assistant cites traces to the document it came from, by address and by hash. Two documents that disagree are both kept, and the answer can say which one printed what.

A citation of a revision that had not been published yet

Labeling with people in the loop

The register: The queue of decisions that people label.
A subject: One decision. Each label on it is a record with a typed source: who gave it, and in what role.
What the engine gives: Labels are kept apart by source, so agreement with a reviewer is never mistaken for being right. A bounded audit sample, read by people, is what the error rate is measured on.

A reversal dated before the decision it reverses

Industries

Six registers, one shape. The checks are plain ones: an order of events that cannot have happened, or one thing listed as two.

Oil and gas

The register problem: Wells, licenses, operators and vessels are listed by several regulators and registries, each with its own identifiers and spellings.
A subject: One asset, a well for example, across its permits, inspections and incident reports.
What the checks and receipts catch: A well cannot be plugged before it is drilled. An order like that is held as a doubt with both documents attached, and the correction that settles it carries a receipt.

Plugged before it was drilled

Shipping and maritime

The register problem: Seafarer certificates are issued by flag states and training centers, and no two of them print a name the same way.
A subject: One seafarer, or one vessel, across every certificate, crew list and register.
What the checks and receipts catch: Sea time served in a rank before the certificate for that rank was issued, and the same vessel sailing under two names.

Sea time in a rank before its certificate

Finance and compliance

The register problem: Sanctions and KYC screening reads one person across watchlists, corporate registries and transaction parties, each with its own transliteration.
A subject: One person, or one company, across all of them.
What the checks and receipts catch: A match arrives with its evidence and a bounded false-match rate, not a bare percentage. A directorship dated before the company was formed is held as a doubt.

A director of a company that did not exist yet

Retail and shopping

The register problem: One product arrives from suppliers, marketplaces and catalogs under different titles and codes, and one customer shops across stores and devices.
A subject: One product, or one customer.
What the checks and receipts catch: Two prices for one SKU on one day, and a return that predates the purchase.

A return dated before the purchase

Healthcare and professional licensing

The register problem: One practitioner is licensed by several state boards and credentialed by several employers, and the boards do not read each other.
A subject: One practitioner across boards, employers and rosters.
What the checks and receipts catch: Discipline in one state and an active license in another, read side by side, each with the board document behind it.

A new license while discipline stands in another state

Sports governance

The proving ground

The register problem: Athletic commissions each publish their own results and their own suspensions, and an athlete competes under several of them.
A subject: One athlete across every commission.
What the checks and receipts catch: A bout dated inside a suspension window, shown with the results sheet and the suspension notice side by side.

A bout inside a suspension window

Registers join into one subject. Its events run left to right in the order the documents date them. The marked event is the one that cannot be true.

How it is used

Bring a register
Published files, in whatever shape they come. We write the schema and the mapper.
Get subjects
One record per subject, each fact with its document, each change with its receipt.
See the measurement
The error rate on your labeled sample, against the bound we hold.

Send a register. See what resolves.

Tell us which registers you read today and what a wrong match costs you. We come back with what we would need to read them.

We reply by email.

We use what you send only to reply. Nothing is added to a mailing list. Privacy policy

Open. Compile. Receipt.

The most dangerous registers are the least joined.

Public is not the same as open.

A register is a description, not a program.

The proving ground

Same four pictures, three schemas

One engine, any register

Pansophia

What a record carries

Measured on public data

Three asks. One record.

For laboratories and agent builders

For keepers of registers

For the person on the record

A fingerprint is what verify computes.

The evidence, written down.

The evidence layer

Press

Coverage

Where it is used

For AI labs and teams that run agents

Industries

How it is used

Bring a register

Get subjects

See the measurement

Send a register. See what resolves.