@urbankitstudio/atlas 0.6.42 → 0.6.44

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md ADDED
@@ -0,0 +1,2243 @@
1
+ # @urbankitstudio/atlas Changelog
2
+
3
+ ## 0.6.44 — 2026-10-25
4
+
5
+ The three confidential-flagged Texas records' `sampleQuery` URLs exclude flagged owners, and the advised
6
+ `where` clause casts the flag column to text. No type changes; `totals` are unchanged.
7
+
8
+ ### The flagged records' sample queries no longer return flagged owners
9
+
10
+ In 0.6.43 the `sampleQuery` of `tx-harris`, `tx-collin` and `tx-ector` searched owner names
11
+ (`UPPER(owner) LIKE UPPER('%SMITH%')`) without the flag column in `outFields` and without an exclusion, so
12
+ running a record's own sample could return an owner the county marks confidential under Texas Tax Code
13
+ §25.025. Each now requests the flag column and ANDs the exclusion:
14
+
15
+ - **Harris** (`tx-harris`): `... AND CAST(confidential_flag AS VARCHAR(255)) NOT LIKE '%Y%'`, outFields gain
16
+ `confidential_flag`.
17
+ - **Collin** (`tx-collin`): `... AND CAST(GIS_DBO_AD_Entity_confidential_ AS VARCHAR(255)) NOT LIKE '%T%'`,
18
+ outFields gain `GIS_DBO_AD_Entity_confidential_`.
19
+ - **Ector** (`tx-ector`): `... AND CAST(conf_flg AS VARCHAR(255)) NOT LIKE '%1%'`, outFields gain `conf_flg`.
20
+
21
+ Measured live 2026-10-25, whole layer (`1=1`, `returnCountOnly`): Harris 1,551,736 rows, 4,830 flagged and
22
+ 5,235 with a NULL flag cell, 1,541,671 kept by the clause; Collin 452,384 rows, 1,598 flagged and 15,003
23
+ NULL, 435,783 kept; Ector 77,042 rows, none flagged and none NULL, 77,042 kept. Both sums are exact, so the
24
+ clause drops every flagged row and every NULL cell. Each new sample answers HTTP 200 with 10 rows, the flag
25
+ column present and no flagged row.
26
+
27
+ `validate-atlas` now fails a record whose endpoint carries `confidentialFlag` and whose `sampleQuery` lacks
28
+ the flag column in `outFields` or the CAST exclusion for each value in its `where`, or whose `where` puts an
29
+ OR before the exclusion at the top level (an OR glued to a number or a bracket, as in `1=1OR 1=0`, counts:
30
+ the server reads it as an OR, and that shape returned all 4,830 flagged Harris rows on 2026-10-25).
31
+
32
+ ### The advised `where` clause casts the column
33
+
34
+ The 0.6.43 README, type comments and changelog advised `<field> NOT LIKE '%<value>%'`. On Ector, whose
35
+ `conf_flg` is an integer, that answers with an ArcGIS error (code 400, inside an HTTP 200 response, so a
36
+ client that checks only the HTTP status misses it). `CAST(<field> AS VARCHAR(255)) NOT LIKE '%<value>%'` runs
37
+ on all three layers with the same counts. A NULL cell still fails it and the row drops out, and a missing
38
+ column makes the query fail, so the SQL route stays fail-closed. The 0.6.43 entry below also gave Collin `<> 'T'`, which contradicted its own one-character
39
+ rule; Collin's `"T"` takes `NOT LIKE '%T%'` like any one-character value. A longer value still uses
40
+ `CAST(<field> AS VARCHAR(255)) <> '<value>'`.
41
+
42
+ The SQL route compares the untrimmed text: `TRIM(<field>)` answers with the same ArcGIS error on all three
43
+ layers, so the "trimmed and upper-cased" comparison belongs in a row filter in code.
44
+ `UPPER(CAST(<field> AS VARCHAR(255)))` runs on all three. Each layer reported its column type on
45
+ 2026-10-25: Harris `confidential_flag` is a string of length 10, Collin `GIS_DBO_AD_Entity_confidential_` a
46
+ string of length 1, Ector `conf_flg` a small integer. Harris's 470,972 blank cells hold spaces (they match
47
+ `LIKE ' %'`) and the clause keeps them. Separately, Harris has 5,235 NULL cells and Collin 15,003, which
48
+ 0.6.43 did not record; the clause drops them.
49
+
50
+ ### The site's owner-search examples for these counties carry the exclusion
51
+
52
+ On the Texas state page, the Harris owner-search examples (the hand-written FAQ and the state notes) now
53
+ request `confidential_flag` and carry the clause. The generated state FAQ does the same for any state whose
54
+ example county is flagged.
55
+
56
+ ### The tarball carries this changelog
57
+
58
+ `CHANGELOG.md` joins the package's published `files`; the 0.6.43 tarball had none.
59
+
60
+ ## 0.6.43 — 2026-10-24
61
+
62
+ One optional endpoint key and three Texas records that carry it (two measured, one a forward guess). No breaking change; `totals` are unchanged.
63
+
64
+ ### `confidentialFlag` on an endpoint record
65
+
66
+ `EndpointRecord.confidentialFlag?: { field: string; values: string[] }` (type `ConfidentialFlag`, exported)
67
+ names the county's own confidential-owner column on a layer and the values that mean "confidential". Texas
68
+ Tax Code §25.025 lets peace officers, judges, family-violence victims and others have their home address
69
+ kept confidential in appraisal records; Harris and Collin publish those accounts with the owner and mark them
70
+ in a flag column beside it. A row whose flag is set must not be shown or passed on: request the column
71
+ whenever you request the owner, drop a row whose cell, trimmed and upper-cased, equals a listed value or
72
+ (for a one-character value) contains it, and treat a row that lacks the column as flagged. The hosted REST
73
+ API, the hosted MCP and the site's free tools withhold such a row as `withheld: "county-redacted"`, exactly
74
+ like a row the county redacted itself, and never serve the flag column.
75
+
76
+ - **Harris County, TX** (`tx-harris`): `{ "field": "confidential_flag", "values": ["Y"] }`. The column is a
77
+ ten-character string with one Y/N per ownership slot. Measured 2026-10-24 over 1,551,736 rows:
78
+ `'Y '` 1,658 and `'YNNNNNNNNN'` 3,172 (4,830 flagged, every Y in slot 1), `'NNNNNNNNNN'` 1,066,977,
79
+ `'N '` 3,721, blank 470,972 (the county's "no flag recorded", not confidential). In a `where` clause:
80
+ `AND confidential_flag NOT LIKE '%Y%'` (superseded in 0.6.44 by the CAST form).
81
+ - **Collin County, TX** (`tx-collin`): `{ "field": "GIS_DBO_AD_Entity_confidential_", "values": ["T"] }`. The
82
+ layer's alias for the column is "Confidential Flag (t/f)". Measured 2026-10-24: `'T'` on 1,598 of 437,381
83
+ rows, `'F'` on the rest. In a `where` clause: `AND GIS_DBO_AD_Entity_confidential_ <> 'T'` (corrected in 0.6.44:
84
+ `AND CAST(GIS_DBO_AD_Entity_confidential_ AS VARCHAR(255)) NOT LIKE '%T%'`).
85
+ - **Ector County, TX** (`tx-ector`): `{ "field": "conf_flg", "values": ["1"] }`, a forward GUESS, not a
86
+ measurement: its `conf_flg` column (a small integer) was 0 on all 77,042 rows (2026-10-24), so no value
87
+ that means "confidential" has been observed; the marker reads a future non-zero cell as flagged. The
88
+ record's `notes` say so.
89
+
90
+ `validate-atlas` requires the field to read as a confidentiality flag (`conf`), not as an owner, mailing,
91
+ situs or id column, and the values to be in their trimmed upper-case form; every registration of one layer
92
+ URL must carry the same marker.
93
+
94
+ ## 0.6.42 — 2026-10-24
95
+
96
+ One county joins, one county is recorded as unreachable, two county records are retired, the county office's
97
+ email is on both answer types, the county record gains a `credit` field that every surface shows, the endpoint
98
+ record gains a `sharedService` marker on four statewide services, and the San Francisco city datasets move
99
+ host. No type is removed or renamed (`credit`, `county_office.email` and `sharedService` are additions), but
100
+ two records lose their endpoint: `wa-thurston` and `sc-charleston` now carry `hasPublicRest: false` and an empty
101
+ `endpoints` array (`coverage.reason: "licence-restricted"`), so a consumer that read either county's
102
+ `endpoints[0]` finds none and a `coverage` answer instead. `totals` move from 499 counties / 497 endpoints / 470
103
+ with an endpoint (0.6.41 moved none) to 501 / 496 / 469: San Francisco and Laurel join (two counties, one
104
+ endpoint), Thurston and Charleston lose theirs (two endpoints, two counties with one).
105
+
106
+ ### New: San Francisco, California
107
+
108
+ `ca-san-francisco` (06075), from SFGIS's daily ArcGIS Online copy of the DataSF dataset "Parcels - Active and
109
+ Retired" (org Zs2aNLFN00jrS4gG, item f81b5f0adb8c4d6a95282bec1ce10378): `Active_Parcels_from_DataSF_pulled_daily_`
110
+ FeatureServer layer 0, 227,695 active parcels on 2026-10-20, refreshed daily, CORS open, pages
111
+ (maxRecordCount 2000, resultOffset accepted). THE LAYER PUBLISHES NO OWNER NAME AND NO OWNER MAILING ADDRESS:
112
+ the Assessor-Recorder says state law bars it from posting ownership online, so the record carries
113
+ `capabilityOverrides.owner_name: not_published` and an `ownerFieldNote`. The parcel number is `blklot` (block
114
+ plus lot, text, letter suffixes possible); the situs is split into `from_address_num`/`to_address_num` and
115
+ `street_name`/`street_type`, so an address lookup composes House Number plus Street Name, and a parcel whose
116
+ range spans several house numbers matches only at its low end. Licence PDDL (the ArcGIS item's and DataSF's).
117
+ `credit: "City and County of San Francisco (DataSF, SFGIS)"`. The `notes` name the fallback, SF Planning's
118
+ PlanningData MapServer layer 23. `relatedPropertyCitySlug: "san-francisco"` links the record to the city's
119
+ permit and violation datasets, whose registry entry now carries `countySlug: "san-francisco-county"`.
120
+
121
+ ### Recorded as unreachable: Laurel County, Kentucky
122
+
123
+ `ky-laurel` (21125) ships with `hasPublicRest: false` and `coverage.reason: "no-public-service"`. The PVA
124
+ publishes parcels only through its qPublic website (laurelpva.com), which has no query service and answers
125
+ automated requests with a browser challenge; the state holds Laurel's parcels "Without Ownership" and has no
126
+ statewide layer yet. The record answers a search for the county on 2026-10-19 with the PVA's site and phone
127
+ number, so `findCounty()` resolves it and `coverageAnswer()` says why there is nothing to query.
128
+
129
+ ### Retired: Thurston County, Washington
130
+
131
+ `wa-thurston` now has no endpoint (`hasPublicRest: false`, `coverage.reason: "licence-restricted"`). The
132
+ county's reply of 2026-10-19 set out its data terms as non-negotiable: term 8 bars making the data available to
133
+ third parties naming Thurston County as the source, and term 10 bars commercial owner mailing lists
134
+ (https://www.thurstoncountywa.gov/map-viewer-disclaimer/disclaimer-data). The endpoint is withheld from the
135
+ atlas; the record keeps its GeoData Center contact and now carries the county's email,
136
+ GeoData@thurstoncountywa.gov, which is where to ask the county to allow it. Owner names were never on the
137
+ county's public layer, so the mailing-only set (Pima AZ, Riverside CA) is what remains. The hosted
138
+ county-access floor moves 53067 to `all-off` the same day, so the hosted API and MCP answer `paused` for it on
139
+ every surface.
140
+
141
+ ### Retired: Charleston County, South Carolina
142
+
143
+ `sc-charleston` (45019) now has no endpoint either (`hasPublicRest: false`, `coverage.reason:
144
+ "licence-restricted"`). Charleston County GIS told us on 2026-10-21 that it "does not authorize this use of our
145
+ data", so the endpoint is withheld from the atlas; the record keeps its GIS contact and carries the county's
146
+ email, WebMasterGIS@charlestoncounty.org, which is where to ask the county to allow it. The hosted county-access
147
+ floor moves 45019 to `all-off`, so the hosted API and MCP answer `paused` for it on every surface.
148
+
149
+ ### The county office's email
150
+
151
+ - `CountyPausedAnswer.county_office` and `CoverageAnswer.county_office` are now
152
+ `{ department, url, email }`, each `string | null`. The email is the record's `contact.email`
153
+ (`AccessRecordLike.contact.email?`, `CoverageRecordLike.contact.email?`); a record without one gives
154
+ `email: null`, the key always present. It is how a person asks the county for the data the answer
155
+ cannot give.
156
+ - `coverageAnswerText()` prints the email on the County office line after the url, so every text surface
157
+ (the hosted MCP, the `uks` CLI 0.1.8) shows it.
158
+ - Under `free_discovery`, `pausedAnswer()` reads the email by the rule the department and url already
159
+ follow: an address at one of the record's endpoint hosts is null, a plain address stays, and a retired
160
+ record (no endpoint) has no host to name, so its email is shown under every preset, `all-off` included.
161
+
162
+ ### `credit` on the county record, shown wherever the data is
163
+
164
+ - `CountyRecord.credit?: string`: the statement the publisher's terms require or request wherever its data
165
+ is shown, verbatim and never shortened. Four classes carry one. The ten Colorado counties on the State's
166
+ `Colorado_Public_Parcels` service (El Paso, Jefferson, Arapahoe, Adams, Douglas, Larimer, Boulder, Weld,
167
+ Mesa, Park) carry "State of Colorado Public parcels were accessed on 2026-10-17 from
168
+ https://gis.colorado.gov/public/rest/services/Address_and_Parcel". `ca-san-francisco` carries "City and
169
+ County of San Francisco (DataSF, SFGIS)". `mo-greene`, whose layer is the City of Springfield's, carries
170
+ the City's "Use of Data" disclaimer in full (https://www.springfieldmo.gov/disclaimer, now the endpoint's
171
+ `licenseUrl`), because the City requires any person, company or agency using data from its open data
172
+ systems to include it wherever the data is distributed. The same text stays in `notes` after `Credit:`.
173
+ `pa-berks` carries "Information is based on geospatial data from the County of Berks." (added 2026-10-24):
174
+ clause 4 of the County's data agreement requires that sentence on every product derived from its data, so
175
+ it goes with every Berks row, file and label sheet, the free single-parcel lookup included.
176
+ - Where it shows: with everything served from the record's layer. The hosted REST API carries it as
177
+ `county.credit` on `/api/atlas/lookup` and `/api/atlas/county`, as `credit` on every `/api/atlas/bulk` row
178
+ the county's layer answered, and as `county.credit` on a covered `/api/atlas/radius` answer. The hosted MCP
179
+ carries it inside `county` on `get_county_endpoint` and `find_county_by_fips` (the whole record), as
180
+ `atlas_match.credit` on `find_parcel_by_address`, per county on `list_counties_in_state`, as `credit` on an
181
+ `enrich_address` answer, and as `county.credit` on `radius_owners`. The site prints "Data credit: ..."
182
+ beside the rows on Parcel Lookup (a pasted credited layer included, and in its PDF), Find My Parcel Number,
183
+ Owner Lookup, the radius notice tool (and a `Data credit` column in its label CSV) and bulk enrichment (and
184
+ a `Data credit` column in its CSV); the ArcGIS to GeoJSON converter writes it as a top-level `credit` member
185
+ of the FeatureCollection; the county page shows it under its own heading. A withheld record carries none,
186
+ since it shows no data; the paid rows still served for a county whose record is withheld from discovery
187
+ carry it unless it names the county's service (Colorado's does, so it is null there). A consumer of this
188
+ package reads `credit` and shows it as is.
189
+ - The validator's cap on `credit` is 600 characters (`CREDIT_MAX` in `scripts/atlas-rules.mjs`), bounding a
190
+ paragraph rather than a sentence, because Springfield's statement is 452 characters and a required statement
191
+ is never truncated.
192
+ - A `credit` holds only WinAnsi characters (printable ASCII, U+00A0..U+00FF and the 27 Windows-1252 additions
193
+ such as curly quotes and dashes; `creditCharacterProblem` in `scripts/atlas-rules.mjs`), the set a PDF's
194
+ standard Helvetica can print, so the statement prints verbatim in every PDF export.
195
+
196
+ ### `sharedService` on the endpoint record
197
+
198
+ `EndpointRecord.sharedService?: true` (added 2026-10-24): one county registers a layer of the service, but the
199
+ record's own notes say the service covers other counties or the whole state, so its other layers may be any
200
+ county's. Four endpoints carry it: Chittenden VT (VCGI's statewide `FS_VCGI_VTPARCELS_WM_NOCACHE_v2`, "covers
201
+ all 14 Vermont counties"), Kauai HI (the State's `ParcelsZoning` MapServer, where Honolulu, Maui and Hawaii
202
+ County parcels are separate layers), Denver CO (`ODC_PROP_PARCELS_A`, a statewide composite) and Anchorage AK
203
+ (`AK_Parcels`, the statewide layer). The site's `findServiceCounty`, which places an unregistered layer of a
204
+ one-county service as that county's, places a layer of a marked service as no county's, and the registry
205
+ validator requires the marker whenever a record's prose says so. The site's free tools still ask the county
206
+ access switch about every county registered on a marked service before reading one of its layers (a pause on
207
+ Chittenden holds VCGI's other layers): the data is at least that county's, though the layer gets no county and
208
+ no credit. Absent is the other value; `false` is never written. The other 178 single-registration services were
209
+ surveyed (their service metadata and layer lists fetched) and none other covers another county.
210
+
211
+ ### San Francisco city datasets: data.sf.gov
212
+
213
+ The two DataSF Socrata datasets in `city-property/san-francisco.json` (building permits `i98e-djp9`, notices
214
+ of violation `nbtm-fbw5`) now point at `data.sf.gov`. The old host, `data.sfgov.org`, answers 301 with
215
+ `Access-Control-Allow-Origin: https://data.sfgov.org`, which a browser refuses, so the site's fetches failed;
216
+ the new host answers 200 with `Access-Control-Allow-Origin: *` (verified 2026-10-20 with `$limit=1` and an
217
+ Origin header on both ids). The registry validators now reject `data.sfgov.org` anywhere in a registry file.
218
+
219
+ ### Package metadata
220
+
221
+ Metadata only. The package author and the license holder now name UrbanKit Studio LLC.
222
+
223
+ ## 0.6.41 — 2026-10-24
224
+
225
+ New Jersey: owner names and owner mailing addresses are withheld for all six counties. Data change only, no
226
+ code or API change, and no county joins or leaves (`totals` do not move).
227
+
228
+ - **All six NJ counties** (Bergen 34003, Camden 34007, Essex 34013, Monmouth 34025, Morris 34027, Ocean 34029)
229
+ now carry `capabilityOverrides.owner_name` and `capabilityOverrides.owner_mailing_address` as `restricted`,
230
+ citing Daniel's Law (N.J.S.A. 56:8-166.1, P.L. 2020, c.125, as amended), which restricts disclosure of
231
+ the home addresses of judicial officers, law enforcement officers, child protective investigators and
232
+ prosecutors, and their household family members. A parcel layer does not identify which parcels belong to
233
+ those persons, so the package leaves both roles out. `ownerFieldNote` is added on Monmouth and extended on
234
+ the other five.
235
+ - The mailing columns (`ST_ADDRESS`, `CITY_STATE`, `ZIP_CODE`) are dropped from `searchFields` and
236
+ `sampleQuery` on Bergen, Camden, Essex, Morris and Ocean. On Monmouth the owner columns (`Owner_Name`,
237
+ `Owner_Street`, `Owner_Csz`) are dropped, and `sampleQuery` now searches `Location` instead of
238
+ `Owner_Name`. Anything that reads these records to build an owner or mailing query for an NJ county now
239
+ finds no such field and a `restricted` status.
240
+ - Monmouth's earlier note described Daniel's Law as covering domestic-violence victims and cited
241
+ P.L. 2021, c.371. Both are corrected: the statute covers the groups above and is cited as above.
242
+ - Consumers of this package (the stdio MCP server `@urbankitstudio/mcp-atlas` among them) pick the change up
243
+ on a dependency bump to 0.6.41.
244
+
245
+ ## 0.6.40 — 2026-10-18
246
+
247
+ Five counties join, three county records move and one sentence is added. No breaking change. `totals` move
248
+ from 494 counties / 494 endpoints / 467 with an endpoint to 499 / 497 / 470. North Carolina now carries all
249
+ 100 of its counties (Hoke and Swain were the two the 0.6.37 release left out), Ohio eleven and Illinois ten.
250
+
251
+ ### New counties
252
+
253
+ - **Rock Island County, IL** (`il-rock-island`, 17161), from the county GIS department's ArcGIS Online
254
+ feature service (services9.arcgis.com, item 9cae8a64ab0e4cea99758f741ca43b3c, "Parcels", owner
255
+ thirty34jb, the county GIS account): `Parcels` FeatureServer layer 0, `parcel_layer`, 65,948 rows, last
256
+ edited 2026-10-01, tax year 2025. The county's own host, gis.rockislandcountyil.gov
257
+ (`Hosted/Parcels/FeatureServer/0`), now answers "Token Required", so the record points at the ArcGIS
258
+ Online copy. Searchable by `parcel_number`, `owner1_name`, `taxbill_name`, `site_address`, `Site_City`
259
+ and `owner1_address1`; the layer also carries the owner and tax-bill mailing addresses, the assessed
260
+ values (EMV, EAV, farm and non-farm land and building), the last sale, class, township, legal
261
+ description, zoning, year built and square footage. 65,763 rows carry a parcel number and every one of
262
+ them an owner name; 58,640 carry a site address. The native spatial reference is Illinois State Plane
263
+ West, feet (WKID 3436), so a point query passes `inSR=4326`; the layer pages, declares
264
+ `maxRecordCount: 2000` and answers with CORS open. `license: "public"`: the item's license text reads
265
+ "For use by the general public", and the county's Data Release Policy is as-is and no-warranty with a
266
+ source credit of Rock Island County GIS, no resale ban and no fee stated. Credit Rock Island County GIS.
267
+ Verified 2026-10-18.
268
+ - **Hoke County, NC** (`nc-hoke`), from the county's ArcGIS Server
269
+ (`maps.hokecounty.org`, `OperationalLayers` MapServer layer 6, "Parcels",
270
+ 29,570 rows). The NC OneMap statewide layer has no owner or address values for
271
+ Hoke. Searchable by `TWN_PIN`, `OWNER_NAME` and `PHY_ADDRES`; it also lists the
272
+ second owner name, the owner's mailing lines (`MAILING_AD`, `MAILING__1`) with
273
+ `CITY`, `STATE_OR_C` and `ZIP_CODE`, calculated acres, `COST_TOTAL` (the layer
274
+ gives no definition) and the deed date. The layer has no site-city or site-ZIP
275
+ column: `CITY`, `STATE_OR_C` and `ZIP_CODE` are the owner's mailing city, state
276
+ and ZIP. It pages and declares `maxRecordCount: 1000`. 95 rows have no owner,
277
+ and one parcel number repeats. `license: "unknown"`: the county publishes no
278
+ license text, only a liability notice on its GIS page.
279
+ - **Swain County, NC** (`nc-swain`), from the county GIS office's ArcGIS Online
280
+ feature service (`Swain_County_Parcels` FeatureServer layer 0,
281
+ `Parcel_Polygons_Public`, 13,554 rows, last edited 2026-09-29). The NC OneMap
282
+ statewide layer has no parcel numbers for Swain. Searchable by `PARCEL_ID`,
283
+ `Name__1` (the owner name, which the layer does not define) and
284
+ `ParcelPropertyAddress`; it also lists the second name, the owner's mailing
285
+ lines and city, state and ZIP (the layer has no site city), acres, land,
286
+ building and total assessed value, and the deed date. 599 parcel numbers repeat
287
+ across rows (798 extra rows, at most 45 under one number) and 46 rows have none.
288
+ `license: "unknown"`: the county item's license text is a warranty disclaimer
289
+ that places no limit on use.
290
+ - **Licking County, OH** (`oh-licking`), from the county's own ArcGIS Server
291
+ (`gis.lickingcounty.gov`, `Auditor/Parcels` MapServer layer 0, 83,782 rows; the
292
+ latest transfer is dated 2026-10-15). Searchable by `Parcel`, `OwnerName` and
293
+ `SiteAddress`; it also lists the site city and ZIP, the four owner mailing
294
+ columns, municipality, township, tax acres, land use, property class, market
295
+ and net total value, year built, and the latest transfer date and amount. 1,189
296
+ rows are shapes with no auditor record, and every parcel number is unique. The
297
+ layer pages and declares `maxRecordCount: 2000`. `corsEnabled: true`: on
298
+ 2026-10-17 the server echoed the request origin on the layer and on `/query`,
299
+ though an earlier probe the same day found no CORS header. `license: "unknown"`:
300
+ the county's GIS downloads page carries a reference-only, no-warranty
301
+ disclaimer and states no limit on redistribution.
302
+ - **Madison County, OH** (`oh-madison`), from the County Engineer's ArcGIS Online
303
+ feature service (`Parcels_Engineer` FeatureServer layer 13,
304
+ `Parcels_Parcels_join`, 25,931 rows, last edited 2026-10-15), which joins the
305
+ Engineer's parcels to the Auditor's. Searchable by `TAXPIN` (the auditor's
306
+ parcel number; `PARCEL_NO` is the Engineer's own sequence number and repeats on
307
+ up to 223 rows, so it is not listed), `NAME` (the owner), and the site address,
308
+ which the layer splits into `Number` and `Street` (the bare street name, with no
309
+ street type). It also lists the site city and ZIP (blank on 40.4% of rows), the
310
+ owner's mailing street, city, state and ZIP (`Owner_Cont`, `Owner_Co_1`,
311
+ `Owner_Co_4`, `Owner_Co_5`; the layer's `Taxpayer_*` columns are a different,
312
+ tax-bill address and are not listed), acres, the transfer date and the tax
313
+ district. 287 rows carry the placeholder parcel numbers 99-99999.999 and
314
+ 88-88888.888 and no owner, and 1,600 other parcel numbers repeat across rows, so
315
+ a parcel-number lookup can return several rows. `license: "unknown"`: the
316
+ parcel item and the county's GIS Data Download hub state no terms, while two
317
+ other items in the same ArcGIS Online organization (the Comprehensive Plan
318
+ layers) carry a notice ending "Reproduction in any form is prohibited without
319
+ the expressed consent of Madison County"; the record's notes quote both.
320
+
321
+ The four records probed on 2026-10-17 carry in their `notes` the share of rows filled for every column the
322
+ atlas lists, read that day, and the extent and county-seat checks (each layer's extent falls inside its
323
+ county, and a point or box at the county seat intersects parcels). `lastVerified` is the probe date, not a
324
+ data date: none of those four layers dates its roll. Each answered one bounded query again on 2026-10-18.
325
+
326
+ ### Records moved
327
+
328
+ - **Fulton County GA (13121) repointed.** The county's on-premises server (gismaps.fultoncountyga.gov,
329
+ PropertyMapViewer layer 11) stopped accepting connections on 2026-10-18. The record now points at Fulton County
330
+ GIS's own ArcGIS Online copy of the tax parcels, `Tax_Parcels/FeatureServer/0` on services1.arcgis.com (item
331
+ e581a072dca9442e884d3682bff03484, owner Fulton_County_GIS, public, last edited 2026-09-22): 373,306 parcels,
332
+ TaxYear 2026, the same nine fields with the same labels, maxRecordCount 1,000, pagination on, CORS open. The
333
+ 42 rows whose owner reads RESTRICTED stay withheld. Use the unsuffixed `Tax_Parcels` service, not the
334
+ `Tax_Parcels_2025` and `Tax_Parcels_2024` snapshots beside it.
335
+ - **Greenville County SC (45045) retired, with the reason.** `hasPublicRest: false`, `endpoints: []`,
336
+ `coverage.reason: "owner-data-sold"`, checked 2026-10-18: the county removed every public map service from
337
+ gcgis.org (the services root lists none; the registered layer answers "Service not found") and sells its tax
338
+ parcels with ownership as countywide vector data on a USB drive, $500 in shapefile format, under a data order
339
+ form and release agreement; `coverage.countyProduct` carries that page. Owner lookups are on the county's
340
+ Real Property Search, and no statewide South Carolina layer carries owner names. The record had been
341
+ `unreachable` since 2026-08-29.
342
+ - **Montgomery County KY (21173) retired, with the reason.** `coverage.reason: "login-required"`, checked
343
+ 2026-10-18: around 2026-10-15 the county's GIS host, the Gateway Area Development District, made the
344
+ `Montgomery_Parcels` FeatureServer private, and the layer now answers "Token Required". The organisation's
345
+ public layers carry no owner names. `contact.url` is the county's site. The county's KRS 61.874 pause on paid
346
+ surfaces is live data from the hosted API and is unchanged.
347
+ - **`countyDownMessage(countyName)`**: the one sentence every surface gives when a county's own GIS server is
348
+ down (the liveness probe observed `down`, the operator paused the county with reason `upstream-broken`, or a
349
+ query failed on the county's side): "Fulton County's GIS server is down on the county's end, so its parcel data
350
+ can't be reached right now. Try again later, or use the county office's own site." Without a name it reads
351
+ "The county's GIS server…". It says nothing about money: whether a call was charged is that call's fact, so each
352
+ surface appends its own charge clause. Exported from the package entry. On the hosted REST API, a lookup on
353
+ a county whose own server the liveness probe currently reads as down is served with this sentence and is
354
+ not charged on any tier (Leo, call 11, 2026-10-18); a reading older than six hours, or none, charges as
355
+ before.
356
+ - **`coverageOverridesPause(record)`**: true for a record with no endpoint whose `coverage.reason` is
357
+ anything but `licence-restricted`. The hosted API's lookup, county and MCP discovery tools answer such a
358
+ county with its `coverage` block even while the county access switch pauses it: the pause exists for a
359
+ county we could serve, and this one has nothing to serve, so the record's own reason is the honest answer
360
+ (Montgomery KY). A licence-restricted retired county (Shelby TN) keeps the paused answer, and a county
361
+ whose endpoint returns is paused again with no edit to the switch. A contradictory record
362
+ (`hasPublicRest: false` beside a non-empty `endpoints` array, a shape `validate-atlas` rejects) keeps the
363
+ pause too, so an override never serves endpoints the record itself lists. Exported from the package entry.
364
+
365
+ ## 0.6.39 — 2026-10-18
366
+
367
+ The paused-county answer. No breaking change, no data change, `totals` unchanged.
368
+
369
+ A county can be in the atlas, with a working endpoint, and still be paused for some uses: its own terms
370
+ forbid commercial use, a state law lets the county require a contract, or the county asked. The hosted
371
+ API now answers for such a county with a `coverage` block whose `status` is `"paused"`, and this release
372
+ exports that shape and the function that builds it:
373
+
374
+ - `ACCESS_FLAGS`: the seven surfaces a county can be paused on, each on its own: `paid_lookup` (REST lookup by a
375
+ paying caller, PAYG or a paid plan), `paid_enrich` (MCP `enrich_address`), `paid_bulk` (REST bulk rows),
376
+ `paid_lists` (REST radius, MCP `radius_owners`), `free_site_lookup` (the site's free lookup tools, and REST
377
+ lookup by an anonymous or free-tier caller), `free_site_lists` (the site's free whole-layer download) and
378
+ `free_discovery` (the county record beyond its identity in county metadata and the MCP discovery tools).
379
+ - `ACCESS_PRESETS` and `PRESET_FLAGS`: `"paid-off"` (the four paid surfaces), `"paid-lists-off"` (`paid_bulk` and
380
+ `paid_lists`; single-parcel lookups still work) and `"all-off"` (all seven; the hosted API and MCP answers
381
+ serve only the record's identity: no endpoint URLs, notes or other fields). `flagsFor(preset)`, `pausedFlags(flags)` and `presetOf(flags)` (`"custom"` for
382
+ any other set) convert.
383
+ - `free_discovery` governs the hosted API and MCP answers only. The endpoint URLs in this package's bundled data
384
+ are governed by the data release: a county whose endpoints must not be published becomes a no-endpoint record
385
+ there (0.6.37 removed `tn-shelby`'s endpoint).
386
+ - `ACCESS_REASONS`: `licence-terms`, `state-law`, `county-request`, `data-quality`, `upstream-broken`, `other`.
387
+ - `accessBlocks(entry, flag)`: whether that surface is paused for the county (no entry: never).
388
+ - `pausedAnswer(entry, record, page)` and the `CountyPausedAnswer` type: `indexed: true`, `status: "paused"`,
389
+ `preset`, `paused` (the paused surfaces), `reason`, `why`, `citation_url`, `since`, `county_office`,
390
+ `free_alternatives` (the free county page, null when the free lookup or discovery is paused) and `request: null`.
391
+ **`CountyPausedAnswer.why` is `string | null`, as is `citation_url`**, although the operator's
392
+ `CountyAccessEntry.why` is a `string`: a consumer that shows `why` must handle null (the hosted site shows a
393
+ generic sentence).
394
+ When `free_discovery` is paused, a `county_office` department or url, the `why` or the `citation_url` that names
395
+ one of the record's endpoint hosts or an ArcGIS REST path (a `rest/services` path segment, so not
396
+ `forest/services`, a `sharing/rest` path, an `arcgis.com/home/item` page, or a Map, Feature, Image, GP,
397
+ VectorTile or Scene server) is null; pass the record with its `endpoints` (`AccessRecordLike.endpoints`). The
398
+ value is read with its percent-escapes decoded, in several forms: plain; folded (NFKC, lower case, the
399
+ characters a reader does not see dropped: format characters such as a soft hyphen or a zero-width space, and
400
+ the other default-ignorables such as a combining grapheme joiner or a variation selector; fullwidth and
401
+ ideographic stops as dots); and each again with its control characters deleted, as the URL parser deletes a
402
+ tab or newline. A url a withheld answer links (the `citation_url`, the office url) is also read as the URL
403
+ parser makes it, the link a browser follows. Two passes over the forms decide it, and either one nulls. The
404
+ first parses each dotted name as a host (in the URL parser's ASCII form, without a port or a trailing dot): it
405
+ names the service when it equals an endpoint host (both without a leading `www.`) or is a name under one
406
+ (`fiscalhub.gis.` beside `gis.`). The second looks for an endpoint host without its `www.` with no ASCII
407
+ letter or digit on either side, so any punctuation, symbol, space, `-` or `_` beside it still names it. An
408
+ endpoint host is matched as an ASCII key, so every endpoint host in the atlas is plain ASCII (no `xn--` label,
409
+ no `%`); an IDN host would need the matcher extended first. A parent host (the
410
+ county's own domain beside its `gis.` server) or a sibling of a plain host (an `assessor.` beside `gis.`)
411
+ stays; it fails closed on a name under a `www.` endpoint's domain (`maps.` beside `www.`) and on a longer name
412
+ that begins with the host. Otherwise the office contact stays: a hub or open-data portal, or a department
413
+ named after its GIS organisation, is how a paused county's office is reached. A value over 2,048 characters,
414
+ before or after folding, is null.
415
+
416
+ Which counties are paused is live data, not bundled: read `GET https://urbankitstudio.com/api/atlas/status`
417
+ (its `paused` list) or `?fips=` for one county. A pause or un-pause therefore needs no release of this package.
418
+
419
+ Fixed: `countySlugFromName` drops an apostrophe (straight `'`, `’` U+2019 or `ʼ` U+02BC) before slugging, so
420
+ `"Prince George's"` gives `prince-georges-county`, the slug the atlas stores, not `prince-george-s-county`, which
421
+ `findCounty` does not find. `requestHint` slugs a county name by the same rule. Every bundled county name now gives
422
+ its stored slug (494 of 494; it was 493). Found through mcp-atlas.
423
+
424
+ ## 0.6.38 — 2026-10-17
425
+
426
+ This release gives one answer for a county the atlas cannot serve. Every record
427
+ with `hasPublicRest: false` now carries a `coverage` block with a reason code,
428
+ one sentence and the date we last checked. `coverageAnswer()` turns that record,
429
+ or the lack of one, into the answer the hosted MCP server, the REST API and the
430
+ `uks` CLI give, with a ready-to-send request for the county. Fifteen counties
431
+ are newly recorded with their reason, and 27 records now say why they are not
432
+ served. The SDK lookups also accept a two-letter state abbreviation. Every
433
+ change is additive. No existing field, export or return value changes.
434
+
435
+ `totals` move from 479 counties and 494 endpoints to 494 counties and 494
436
+ endpoints. 467 counties have a verified endpoint, as before.
437
+
438
+ ### The coverage answer
439
+
440
+ - `CountyRecord.coverage?: CountyCoverage | null` holds `reason`, one of
441
+ `COVERAGE_REASONS` (`owner-data-sold`, `no-public-service`, `login-required`,
442
+ `licence-restricted`, `attributes-missing`, `owner-blank`, `service-down`,
443
+ `data-stale`, `technical`), plus `summary`, one sentence of at most 280 characters written
444
+ to be shown as is, and `checked`, the YYYY-MM-DD of the last check. Two keys
445
+ are optional. `recheck` is the YYYY-MM-DD of the next planned check (Columbia
446
+ PA carries one). `countyProduct` names the county's own paid channel, and only
447
+ with `owner-data-sold`, never a reseller.
448
+ - `licence-restricted` (additive) means the county publishes a parcel
449
+ service, but its terms do not allow us to serve it. Shelby TN carries it.
450
+ - `countyHasEndpoint(record)` is the one "is this county served" rule:
451
+ `hasPublicRest` is true and at least one endpoint is registered.
452
+ - `coverageAnswer(record, ref)` returns null for a served county. Otherwise it
453
+ returns a `CoverageAnswer` with `indexed: false` and a `status`.
454
+ `"not-in-atlas"` means the atlas has no record, and says nothing more, since a
455
+ misspelled name lands there too. `"no-public-data"` means the atlas records
456
+ the county and it publishes nothing we can serve. The block also carries
457
+ `reason`, `why`, `checked`, `recheck`, `county_office` (from the record's
458
+ `contact`), `county_product`, `request`, and `message`, which is the whole
459
+ answer as text.
460
+ - `coverageAnswerText(block)` renders that text, one point per line, and it is
461
+ what every text surface prints verbatim. `NOT_IN_ATLAS_WHY` and
462
+ `NO_PUBLIC_DATA_WHY` are its two fallback sentences.
463
+ - `requestHint(ref, status?)` builds the `request` block: a body for
464
+ `POST https://urbankitstudio.com/api/atlas/request` (`COUNTY_REQUEST_URL`) and
465
+ the hosted `request_county` tool, with `COUNTY_REQUEST_MESSAGE` or
466
+ `COUNTY_RECHECK_REQUEST_MESSAGE`. `ref` is
467
+ `{ state, county?, fips? }`. The state can be an abbreviation, a name, a slug
468
+ or a two-digit FIPS code, and any spelling of one county gives the same body.
469
+ It returns null when the ref names no county the endpoint accepts (no valid
470
+ FIPS code, and a state or county with nothing to slug).
471
+ - All of these are pure, with no fetch, no clock and no randomness, so one
472
+ record and ref give the same answer on every surface. The hosted MCP server,
473
+ the REST API and the `uks` CLI 0.1.4 run copies of this same module.
474
+ - Types exported: `CoverageReason`, `CoverageStatus`, `CountyCoverage`,
475
+ `CountyProduct`, `CoverageAnswer`, `CoverageTextInput`, `CoverageRecordLike`,
476
+ `EndpointPresenceLike`, `CountyRequestRef`, `CountyRequestHint`.
477
+ - On the hosted REST API, a plan-key lookup that lands on a county we do not
478
+ serve no longer draws from the key's quota: the draw is given back, and the
479
+ answer says "not billed" only when the credit landed.
480
+
481
+ ### Newly recorded counties (15, no endpoint)
482
+
483
+ - `owner-data-sold`: Lancaster PA (42071), and Kenton (21117), Campbell (21037)
484
+ and Pendleton (21191) KY, whose parcels LINK-GIS publishes and sells.
485
+ - `attributes-missing`: Columbia PA (42037), Trimble (21223) and Hardin (21093)
486
+ KY, Pueblo CO (08101), Pierce WA (53053), Loudoun VA (51107) and Forsyth GA
487
+ (13117).
488
+ - `no-public-service`: Knox (47093) and Rutherford (47149) TN, which
489
+ Tennessee's statewide parcel data leaves out.
490
+ - `data-stale`: Williamson TN (47187), which the statewide data also leaves
491
+ out, and whose only public parcel layer is a copy from 2003.
492
+ - `technical`: Scott KY (21209).
493
+
494
+ ### Coverage blocks on existing no-endpoint records (12)
495
+
496
+ Winnebago IL, Casey KY, Iron MI, Hillsborough and Rockingham NH, Dutchess NY,
497
+ Garfield OK, Providence RI, Webb TX, Jackson MO, Montgomery OH and Shelby TN.
498
+ The two New Hampshire records are also corrected. NH GRANIT's statewide parcel
499
+ mosaic is public, with no login, but it has no owner field, so both are
500
+ `attributes-missing`. Shelby TN is `licence-restricted`, because ReGIS
501
+ publishes a parcel service whose terms bar public display without its
502
+ written consent.
503
+
504
+ - tn-shelby: notes no longer link the county's ArcGIS service. They cite the
505
+ county's ReGIS page (https://www.shelbycountytn.gov/467/REGIS-Regional-GIS)
506
+ and keep ReGIS's quoted terms, and no field of the record names the service's
507
+ host or path.
508
+
509
+ ### State abbreviations in the SDK lookups (additive)
510
+
511
+ `findState`, `findCounty` and `listCountiesByState` still look a state up by
512
+ its slug, and a two-letter abbreviation is also accepted, in either case:
513
+ `findCounty("IL", "kane-county")` finds Kane. The 50 states and DC read as
514
+ their slug by the rule the hosted MCP server and the REST API use
515
+ (`stateSlugFromAbbrev` in the coverage module). `"pr"`, `"gu"` and any other
516
+ input are read as a slug unchanged, so they find nothing, as before. Slugs stay
517
+ exact: `findState("Illinois")` is still undefined.
518
+
519
+ ## 0.6.37 — 2026-10-17
520
+
521
+ This release adds 83 North Carolina counties, all from the NC OneMap parcel
522
+ compilation. Hoke and Swain are left out. The statewide layer carries no owner
523
+ or address values for Hoke and no parcel numbers for Swain. The release also
524
+ adds Union County OH, from the county auditor's service. Martin NC and Union OH
525
+ were the demand triggers. This is a data release with no API change.
526
+
527
+ `totals` move from 395 counties and 413 endpoints to 479 counties and 494
528
+ endpoints, and 467 counties have a verified endpoint. Four records that pointed
529
+ at another state's layer are corrected, and two of them now have no endpoint.
530
+ Shelby TN's endpoint is withheld (see below). North Carolina now carries 98 of
531
+ its 100 counties, and Ohio carries 9.
532
+
533
+ ### New counties
534
+
535
+ - **83 North Carolina counties**, all on NC OneMap's statewide parcel layer
536
+ (`NC1Map_Parcels` MapServer layer 1 on services.nconemap.gov, 5,938,900 rows
537
+ across the 100 counties), in the same shape as the ten counties already
538
+ registered on it (Forsyth, Guilford, Cumberland, Montgomery, New Hanover,
539
+ Iredell, Union, Onslow, Johnston, Brunswick): the same URL, scoped per county
540
+ with `scopeWhere: "cntyname = '<County>'"`, and no `supportsPagination` or
541
+ `maxRecordCount` key (the layer pages, 5,000 rows per request). Each
542
+ `sampleQuery` is a county-scoped `parno` equality (2.0 to 2.8 s on release
543
+ day). Chowan's rows appear doubled: all eight parcel numbers checked match two
544
+ rows each, so its sample returns two.
545
+ - Columns are listed from coverage measured on release day, counting a value of
546
+ only spaces as blank: `siteadd` (searchable), `scity`, `szip`, `parval`,
547
+ `gisacres` and `parusedesc` where at least 50% filled. Seven counties have no
548
+ site address in the statewide layer (Avery, Bladen, Cabarrus, Franklin,
549
+ Orange, Perquimans, Richmond) and carry a reviewed
550
+ `situs_address: not_published` record; search them by owner or parcel number.
551
+ - Mailing columns (`mailadd`, `mcity`, `mstate`, `mzip`) are listed where
552
+ `mailadd` is at least 80% filled and a 25-row sample was street-like on at
553
+ least 90% of rows (four counties at 22 of 25 passed a 50-row resample from a
554
+ later objectid window): 73 counties. The other ten carry a reviewed
555
+ `owner_mailing_address: not_published` record that says why.
556
+ - `license: "public"` with the NC OneMap home page as `licenseUrl`, as on the ten
557
+ counties already registered; the `notes` cite NC OneMap's terms page (a free
558
+ and unrestricted use policy, no warranty), credit the county tax assessor's
559
+ office and NC OneMap (NC CGIA), give each county's measured fill rates, and say
560
+ that no field dates a county's roll: `transfdate` is the state's transform
561
+ date, and Hyde's (2025-03-25) and Clay's (2025-04-17) are 18 months old or
562
+ more.
563
+ - **Union County, Ohio**, from the county's own ArcGIS Server
564
+ (`ParcelAllSaleinfoBuildingInfo` MapServer layer 1, 40,059 rows; sales through
565
+ 2026-09-21). Searchable by `ParcelNo`, `DeedOwner` and `Address`; it also lists
566
+ the site city, `ZipCode` (which holds "CITY OH ZIP", so it is labelled as the
567
+ site city, state and ZIP), the three mailing lines (the third is spelled
568
+ `Mailaddress3` on the layer), appraised value, acres, land use, and sale date
569
+ and price. The layer pages and declares `maxRecordCount: 2000`. 850 rows have
570
+ no owner (90 of them no parcel number either), and 1,045 parcel numbers
571
+ repeat across rows, so a parcel-number lookup can return several rows.
572
+ `license: "unknown"`: the county publishes no license text; its GIS data page
573
+ (the `licenseUrl`) offers the parcels as a free shapefile.
574
+
575
+ ### Corrected records (another state's layer)
576
+
577
+ An extent sweep found four records whose layer belongs to a same-named county in
578
+ another state. Each layer was live and returned real owner rows, so liveness
579
+ checks stayed green.
580
+
581
+ - **Jackson County, Missouri** (`mo-jackson`) pointed at Jackson County,
582
+ Michigan (`gis.mijackson.org`). The county's own current layer
583
+ (`jcgis.jacksongov.org`, `ParcelsAscendRelate` FeatureServer layer 1, 311,163
584
+ parcels) carries parcel ids only, so the record now has no endpoint
585
+ (`hasPublicRest: false`), as Dutchess NY does.
586
+ - **Montgomery County, Ohio** (`oh-montgomery`) pointed at Montgomery County, New
587
+ York. The Auditor's own layer (`gis.mcohio.org`, `AUDGIS_B1` MapServer layer 7)
588
+ carries owners, but only under qualified join names such as
589
+ `SDE.WEB_CAMA.OWNER_NAME1`, which the registry's field-name rule does not
590
+ accept, so the record has no endpoint until it does.
591
+ - **Summit County, Ohio** (`oh-summit`) pointed at Summit County, Utah. It now
592
+ uses the county's own `parcels_web_GEODATA_Tax_Parcels` FeatureServer layer 0
593
+ on `scgis.summitoh.net` (265,902 parcels): owner `ownernme1`, site address
594
+ `siteaddress`, mailing `pstladdress`, `pstlcity`, `pstlstate` and `pstlzip5`,
595
+ and the current market value. It declares `maxRecordCount: 3000`;
596
+ `license: "unknown"`, because neither the layer nor its items publish licence
597
+ text.
598
+ - **Fayette County, Kentucky** (`ky-fayette`) pointed at the Fairbanks North Star
599
+ Borough, Alaska. It now uses LFUCG's own `Parcel` FeatureServer layer 0
600
+ (114,433 parcels: `PVANUM`, site `ADDRESS`, unit, property class, acres). That
601
+ layer publishes no owner name, so `owner_name` carries a reviewed
602
+ `not_published` record on an `endpoint_schema` basis, as Jefferson, Madison,
603
+ Oldham and Henderson do.
604
+ LFUCG's Terms of Use apply to the layer; the notes quote their indemnity
605
+ clause (section VI.B) in full.
606
+
607
+ ### Withheld: Shelby County, Tennessee
608
+
609
+ `tn-shelby` now has no endpoint (`hasPublicRest: false`). The ReGIS parcel
610
+ service's item information says "This data may NOT be displayed publicly, sold
611
+ or transferred to any other entity without the written consent of ReGIS", so the
612
+ endpoint is withheld from the atlas pending that consent. The server also
613
+ requires unsafe legacy TLS renegotiation, which standard clients refuse. The
614
+ record keeps its ReGIS contact.
615
+
616
+ ### Credit lines
617
+
618
+ The registry has no attribution field yet, so each credit is a sentence at the
619
+ end of the record's `notes`, starting `Credit:`. The README's new Attribution
620
+ section lists the same credits.
621
+
622
+ - The ten Colorado counties on the State's `Colorado_Public_Parcels` service (El
623
+ Paso, Jefferson, Arapahoe, Adams, Douglas, Larimer, Boulder, Weld, Mesa, Park)
624
+ carry the attribution statement the service's copyright text names as
625
+ sufficient: "State of Colorado Public parcels were accessed on 2026-10-17 from
626
+ https://gis.colorado.gov/public/rest/services/Address_and_Parcel".
627
+ - Montgomery MD and Prince George's MD credit their county's Planning Department
628
+ (M-NCPPC), which asks for attribution.
629
+ - Wake NC (CC BY 4.0), Sonoma CA (CC BY-SA 3.0 US, "County of Sonoma"),
630
+ Sacramento CA and the District of Columbia (CC BY 4.0) credit the publisher
631
+ their licence names.
632
+ - Union OH's notes quote the county's GIS disclaimer (prepared for tax
633
+ appraisal, no liability, no limit on use).
634
+ - Montgomery MD's notes explain its 4.6-degree extent: 24 of 351,954 points share
635
+ one placeholder coordinate in West Virginia. The layer is the county's own, so
636
+ it takes no county scope.
637
+
638
+ ## 0.6.36 — 2026-10-14
639
+
640
+ City search only. No breaking change, `totals` unchanged.
641
+
642
+ ### Street typed with a house number (Los Angeles violations, Fort Worth permits, Scottsdale violations)
643
+
644
+ These tables keep the house number and the street name in separate columns. A typed address now
645
+ follows the same street rules as Dallas-style tables, with the number kept as its own clause:
646
+
647
+ - A spelled-out direction with its type is the street: `2600 South Blvd` and `100 North Ave` search
648
+ `South` and `North` (they searched `Blvd` and `Ave`).
649
+ - A lettered street keeps its letter: `1700 E Q St`, `100 Q St` and `100 N A St` search `Q`, `Q` and
650
+ `A`. A lone direction letter is a lettered street only in the two-token form before `St` or `Ave`
651
+ (`100 E St` searches `E`); with a direction before it (`100 N E Blvd`) the letter is the street.
652
+ - An abbreviated direction (`N`, `E`, `NE`) plus any other street type is a direction and a street:
653
+ `100 N Court` and `100 E Blvd` search `Court` and `Blvd`. A spelled-out direction is the street name
654
+ (`100 North Court` searches `North`).
655
+ - A number plus one stray token with no street type (`100 %`, `100 E`, `100 St`) is searched as typed,
656
+ never as a bare `%` or `E`. The name columns hold the bare street name, so that search finds no
657
+ rows rather than unrelated rows at that number.
658
+
659
+ ## 0.6.35 — 2026-10-14
660
+
661
+ City search normalization and one new county. No breaking change; one additive
662
+ optional field on the dataset record types (`houseNumberSpaceRuns`, below).
663
+ `totals` move to 395 counties and 413 endpoints (386 with a verified endpoint).
664
+
665
+ ### New county
666
+
667
+ - **Montgomery TN (47125)**, parcel id and situs address only: the county's
668
+ public parcel layer carries no owner or mailing fields, so owner lookups
669
+ return nothing there.
670
+
671
+ ### City search
672
+
673
+ `buildCityQuery` (and the hosted
674
+ `get_city_datasets` `query_url`) now build a more tolerant street match, read
675
+ live on 2026-10-14 against the five datasets that missed.
676
+
677
+ - **Double-spaced addresses (Orlando only).** A dataset opts in with the new
678
+ optional registry field `houseNumberSpaceRuns` (the longest run of spaces to
679
+ try, 1 to 8), set only where a table was measured to store such a run:
680
+ Orlando code cases, permits and zoning cases, at 4. For those, when a typed
681
+ street starts with a house number the query ORs the typed text with one to
682
+ four spaces after the number (`4274 Tatum St` also tries `4274 Tatum St`
683
+ through four spaces), and with the typed text itself when that has a longer
684
+ run or a tab, in parentheses so the dataset scope still applies. Every other
685
+ dataset keeps the one typed literal, exactly as before: the OR group measured
686
+ 1.6 to 1.9x slower on Chicago violations and Orlando permits, which is not
687
+ worth paying where the table has no such run. A group whose encoded length
688
+ would pass about 3,000 bytes falls back to the single literal. Spaces only: a word cannot
689
+ come between the number and the street, so `4274 Tatum St` does not find
690
+ `4274 OLD TATUM ST` and `12 Main` does not find `12 Old Main`. The bound
691
+ is the longest run measured upstream on 2026-10-14, with one step of
692
+ headroom: three in Orlando permits and zoning cases, two in code cases. Orlando
693
+ stores a run there on 99,433 of 146,669 code-case addresses, 55,657 of
694
+ 1,105,323 permit addresses and 221 of 407 zoning-case addresses, so a typed
695
+ full address returned nothing and agents were told to search the street name
696
+ alone. As before, the house number is not anchored on its left: `12 Main` also
697
+ finds `3312 MAIN`. (140,344 code-case addresses carry a double space
698
+ somewhere, mostly before the city name; 99,433 have the run right after the
699
+ number.)
700
+ - **Split-street datasets, street typed without a house number.** Los Angeles
701
+ violations (`stno`/`stname`), Fort Worth permits (`Addr_No`/`Street_Name`) and
702
+ Scottsdale violations (`StreetNumber`/`StreetName`) keep the bare name in the
703
+ address column, so `Funtier Ct` matched as typed found nothing while `Funtier`
704
+ found 87 permits. The name-only path now takes one leading direction and one
705
+ trailing street type off, as the number-plus-name path already did. A trim that
706
+ would leave a lone letter, only a direction or street type, or no letter or
707
+ digit (`N St`, `E St`, `N E St`, a typed `%`) is not applied: the street is
708
+ matched as typed. The derived street name an agent query sends on a Dallas-style
709
+ table (`streetForDataset`) follows the same rule: `N St`, `N Q St` and
710
+ `N % St` are sent as typed, not as `St`, `Q` or `%`, and `100 N St` sends `N St`
711
+ (the name column never holds a house number). A street whose name is itself a
712
+ type or direction word keeps its strip when a real type follows it: `100 Court St`
713
+ sends `Court`, `Plaza Dr` sends `Plaza`. A spelled-out direction with its type
714
+ (`2600 South Blvd`, `100 North Ave`, `South St`) is the street South or North
715
+ and sends `South` or `North`. That holds on tables with no number column and a
716
+ type column (Dallas violations, Boston, San Francisco) for every input, and on
717
+ tables with a number column (Los Angeles, Fort Worth, Scottsdale) only for a
718
+ street typed without a house number; a table with no type column (Chicago
719
+ permits) keeps `100 North St` as `North St`. `N St` and `NE St` stay as typed. A house number
720
+ plus one unsafe token (`100 %`, `100 _`, `100 E`, `100 St`) is sent as typed,
721
+ never as the bare token.
722
+ - **Linear parsing.** The house-number parsers no longer backtrack on a long run
723
+ of digits or of spaces followed by a newline.
724
+ - **Ordinal street names.** A first token like `88th` or `3rd` is no longer read
725
+ as a house number with a glued unit, so `88th St` typed without a number is a
726
+ street name on the split datasets and `3rd Ave` is not cut to `Ave` on Dallas
727
+ violations.
728
+ - Dataset `coverage` and `queryNote` text for these datasets states the new
729
+ behaviour; the "search by street name alone" workaround is gone.
730
+
731
+ ## 0.6.34 — 2026-10-14
732
+
733
+ Mississippi statewide: 80 new counties and a second endpoint for Hinds from the
734
+ MARIS/MDEQ 2024 cadastral layers. These are the first records to declare
735
+ `supportsPagination: false` with `maxRecordCount` 2000 and a county
736
+ `scopeWhere`. Data release, no API change. `totals` move from 314 counties and
737
+ 331 endpoints to 394 and 412 (385 with a verified endpoint). Mississippi is the
738
+ first state with every county on record, 82 of 82.
739
+
740
+ ### New counties
741
+
742
+ - **80 Mississippi counties**, each on one of the two MARIS/MDEQ 2024 statewide
743
+ cadastral MapServer layers on gis.mississippi.edu (East: 44 counties,
744
+ 1,127,050 parcels; West: 38 counties, 878,999 parcels), scoped per county with
745
+ `scopeWhere: "STCNTYFIPS = '28xxx'"`. Both layers refuse pagination
746
+ (`supportsPagination: false`, `maxRecordCount: 2000`; read `offersPagination()`
747
+ before adding either parameter), answer at most 2,000 rows per request with
748
+ `exceededTransferLimit` on a cut, and scan the whole layer on any predicate but
749
+ an FID lookup (1.3 s to more than 20 s, more on the first call after hours idle). Each
750
+ `sampleQuery` is a FIPS-scoped FID equality with no pagination parameter; a
751
+ FID read answers in under a second where a PARNO equality scans the layer.
752
+ - Mailing columns (MAILADD1, MCITY1, MSTATE1, MZIP1) are listed only where
753
+ MAILADD1 is at least 80% populated and a sample read on release day was
754
+ street-like on at least 90% of rows: twelve counties (Clarke, Covington,
755
+ Forrest, Harrison, Jackson, Jones, Lowndes, Oktibbeha, Tishomingo and Wilkinson
756
+ on a 25-row sample; Adams and Union on a 50-row resample from a second FID
757
+ window after the first sample landed at 22 of 25). The other 68 carry a
758
+ reviewed `owner_mailing_address: not_published` record that says why.
759
+ - `license: "unknown"` with the MARIS cadastral data disclaimer as `licenseUrl`.
760
+ The data is public, free and keyless, and no explicit license is stated. The
761
+ `notes` credit the county tax assessors and MARIS/MDEQ; `dataVintage` names
762
+ the 2024 release (TAXYEAR reads 2024 on every row).
763
+
764
+ ### Changed
765
+
766
+ - **Hinds MS** keeps its own-host primary (it paginates) and gains the
767
+ statewide West layer as a second endpoint; both read PARNO. **DeSoto MS**
768
+ keeps its single own-host endpoint: its primary reads PIN and the statewide
769
+ layer PARNO, and a county whose endpoints disagree on the parcel-id column
770
+ could hold no protected-parcel notice entry. Both primaries' `corsEnabled`
771
+ resolves from `null` to `true`, measured with an Origin header. Hinds gains a
772
+ reviewed `owner_mailing_address: not_published` record: the mailing columns
773
+ are empty on both of its layers.
774
+
775
+ ## 0.6.33 — 2026-10-14
776
+
777
+ City data. No API change, no breaking change, `totals` unchanged. The city
778
+ registries grow from 20 to **30 cities** with code-violation or building-permit
779
+ datasets and from 17 to **19** with zoning-case dockets. Every endpoint, address
780
+ column, date column and scope filter was re-read live on 2026-10-14, and each
781
+ new dataset carries `latestRecordDate`, the newest record seen that day.
782
+
783
+ ### New code-violation and building-permit cities
784
+
785
+ - **Los Angeles CA**: LADBS open code-enforcement cases (`u82d-eh7z`; closed
786
+ cases sit in a separate dataset that is not searched) and building permits
787
+ issued since 2020 (`pi9x-tg5x`). The violations table splits the street, so
788
+ a typed "number name" is matched on `stno` and `stname`.
789
+ - **Fort Worth TX**: code violations and development permits (ArcGIS tables).
790
+ The permits table's one-line address column is empty upstream, so permits
791
+ match the street name on `Street_Name` and the house number on `Addr_No`.
792
+ Trade permits are included.
793
+ - **Mesa AZ**: code-compliance cases and issued building permits (Socrata).
794
+ - **Minneapolis MN**: code violations and permits (ArcGIS).
795
+ - **Wichita KS**: MABCD code cases and permits (the city's own ArcGIS server).
796
+ The permits layer is scoped to `Jurisdiction='Wichita'`. The county name is
797
+ Sedgwick County, which the atlas does not carry, so no `countySlug`.
798
+ - **Cleveland OH**: Building and Housing violation notices and building permits
799
+ (ArcGIS), each with its Accela Citizen Access link.
800
+ - **New Orleans LA**: code-enforcement cases and permits (Socrata, CC0).
801
+ - **Orlando FL**: code-enforcement cases scoped to the housing, lot, zoning,
802
+ abatement and commercial case types, and permit applications (Socrata).
803
+ - **Cincinnati OH**: code-enforcement complaints scoped to
804
+ `display_agency like 'Cincinnati%'` (the table is Hamilton County-wide) and
805
+ building permits (Socrata).
806
+ - **Scottsdale AZ**: closed code complaints since 2021-09 (split street, matched
807
+ on `StreetNumber` and `StreetName`) and building permits (the city's own
808
+ ArcGIS server).
809
+
810
+ ### New zoning-case cities
811
+
812
+ - **Orlando FL**: Planning Applications scoped to the Zoning application type.
813
+ Existing and proposed zoning live in the comments prose, not in columns.
814
+ - **Scottsdale AZ**: the ZN (rezoning) series of the planning-case table.
815
+ Thin: about 6 rezonings since 2025-04.
816
+
817
+ ### County links
818
+
819
+ `relatedPropertyCitySlug` now names a city for Los Angeles CA, Tarrant TX,
820
+ Hennepin MN, Cuyahoga OH, Hamilton OH, Orleans LA and Orange FL counties;
821
+ `relatedZoningCitySlug` for Los Angeles CA, Tarrant TX, Orleans LA, Orange FL
822
+ and Maricopa AZ (Scottsdale; Maricopa's property link stays Phoenix).
823
+
824
+ ### Disclosures
825
+
826
+ - The hosted MCP `get_city_datasets` now returns `zoning_source` whenever the
827
+ zoning registry carries a `currentRecordsNote`, not only when its source URL
828
+ differs from the property source, so Orlando's and Scottsdale's zoning notes
829
+ reach agents on the default call. Additive: no field renamed.
830
+ - Dataset `coverage` now states the measured search limits: Orlando's
831
+ double-spaced addresses (search by street name alone), the split-street
832
+ datasets (LA violations, Fort Worth permits, Scottsdale violations: a street
833
+ typed without a house number is matched as typed, so a suffix finds nothing),
834
+ Wichita's null-jurisdiction permits, Cincinnati's departments with their
835
+ shares, and Scottsdale's sparse Zoning column.
836
+ - Types: `CitySource` (the zoning record's `source`) gains the optional
837
+ `currentRecordsNote` its data already carried. Additive.
838
+
839
+ ## 0.6.32 — 2026-10-14
840
+
841
+ New counties — Berks PA (42011), Bedford PA (42009), Salem city VA (51775),
842
+ Nueces TX (48355), Hidalgo TX (48215), Brazos TX (48041), Ector TX (48135) and
843
+ Webb TX (48479, a researched stub with no public REST). No API change, no
844
+ breaking change. `totals` move from 306 counties / 324 endpoints (298 with a
845
+ verified endpoint) to **314 counties / 331 endpoints** (305). Callers asked for
846
+ all eight.
847
+
848
+ ### New counties
849
+
850
+ - **Berks PA (42011)**, with a verified endpoint on PASDA, the Penn State
851
+ clearinghouse that hosts the county's data (`BerksCounty` MapServer layer 6,
852
+ 156,838 parcels): owner, situs, owner mailing address and the 14-digit parcel
853
+ id are searchable, and the layer answers cross-origin requests. The layer is a
854
+ January 2026 snapshot, and PASDA republishes under a new vintage stamp, so the
855
+ layer name or index may move; the `notes` say how to find it again.
856
+ - **Bedford PA (42009)**, with a verified endpoint on PASDA (`BedfordCounty`
857
+ MapServer layer 0, 34,225 parcels). The layer is the July 2023 vintage and
858
+ situs is filled on only about 26 percent of rows, so an address search misses
859
+ most parcels; owner, parcel id and a point lookup are the dependable paths. The
860
+ `notes` record both.
861
+ - **Salem city VA (51775)**, an independent city, with a verified endpoint on the
862
+ city's own ArcGIS Online (`TaxParcels` FeatureServer layer 5, 10,765 parcels,
863
+ edited the day before verification). The mailing street is `owner_address2`;
864
+ `owner_address1` is a care-of line on about 8 percent of rows.
865
+ - **Nueces TX (48355)**, with a verified endpoint on the ArcGIS Online
866
+ organisation of the appraisal district's GIS vendor (`NuecesCADWebService`
867
+ FeatureServer layer 0, 157,404 parcels, edited on the day of verification).
868
+ The site address is split across `situs_num` and `situs_street`, so an address
869
+ search runs on the street word; the `notes` say so.
870
+ - **Hidalgo TX (48215)**, with a verified endpoint on the regional 911 district's
871
+ server (`RGV911_Web_Apps_MIL1` MapServer layer 7, 334,098 parcels). Situs is
872
+ one full string; the mailing column is the street line only, with no mailing
873
+ city, state or ZIP.
874
+ - **Brazos TX (48041)**, with a verified endpoint on the City of Bryan's server
875
+ (`BRAZOS_COUNTY_CERTIFIED_PARCELS` MapServer layer 0, the 2025 certified roll,
876
+ 75,652 parcels, whole county). Empty source fields were written as the text
877
+ `NULL`, and most `SitusAddr` values end in ` NULL`; the data is shipped as
878
+ published and the `notes` say how to query around it.
879
+ - **Ector TX (48135)**, with a verified endpoint on the appraisal-software
880
+ vendor's server (`EctorCounty_Basemap` MapServer layer 0, 77,042 parcels).
881
+ The layer publishes NO situs column: owner, a one-line mailing address and the
882
+ PIN are searchable, and an address lookup has to go through a geocoded point.
883
+ - **Webb TX (48479)**, no public endpoint: `hasPublicRest: false` and an empty
884
+ `endpoints` array. Webb CAD's map host answers with a login page, and the only
885
+ public Webb layers on ArcGIS Online are partial. The `notes` record the
886
+ evidence as of 2026-10-14.
887
+
888
+ ## 0.6.31 — 2026-10-12
889
+
890
+ City datasets. New export, no breaking change, `totals` unchanged. The package
891
+ now bundles the two city registries urbankitstudio.com serves at
892
+ `/data/city-property/` and `/data/city-zoning/`: 20 cities with verified
893
+ code-violation or building-permit datasets and 17 with zoning-case dockets
894
+ (`data/city-property/`, `data/city-zoning/`), and describes how to query each
895
+ one the way the site does.
896
+
897
+ - `listCities(filter?)`, `findCity(ref, state?)` (slug, name, "Name, ST", the
898
+ NYC and DC aliases; ambiguity is a list), `getCityProperty(slug)`,
899
+ `getCityZoning(slug)`, `cityDatasets(slug, kind?)`, `citiesForCounty(stateSlug,
900
+ countySlug)`, `cityPath(slug, kind)`.
901
+ - `buildCityQuery(dataset, street)` returns the exact URL the site sends for a
902
+ dataset and a typed street (Socrata, ArcGIS, CKAN or CARTO), with the site's
903
+ quote-doubling guard; null when the lookup needs a geocoded point or a parcel
904
+ id first, in which case `dataset.queryTemplate` carries the request with a
905
+ `{LON},{LAT}` placeholder. Address lookups carry no template on purpose: a
906
+ street must be escaped for the provider's SQL, and only `buildCityQuery` does
907
+ that. LIKE wildcards (`%`, `_`) in a street stay active, as on the site; each
908
+ `queryNote` says so.
909
+ - Each `CityDatasetDescriptor` carries the provider, endpoint, lookup mode,
910
+ address and date columns, the scope `where` that is part of the dataset's
911
+ definition, the field map, status buckets, coverage, a freshness block, the
912
+ point template where one applies, a newest-first `sampleQuery` and a `queryNote`.
913
+ - County records' `relatedZoningCitySlug` and `relatedPropertyCitySlug` now
914
+ resolve inside the package through `getCityZoning` / `getCityProperty`.
915
+ - Split addresses. A dataset whose table keeps the house number in its own
916
+ column (`addressParts`: Dallas violations, Boston, Chicago, NYC and San
917
+ Francisco, plus the Boston and Indianapolis dockets) is matched on the street
918
+ NAME: `buildCityQuery` takes the typed house number off, then a leading
919
+ directional when the table has a direction column, then one trailing street
920
+ type when it has a type column (`streetForDataset`, `splitAddressStreetName`;
921
+ the roles follow the column order of `addressParts`). A name column holds
922
+ "MARILLA" while "1500" and "ST" sit elsewhere, so the typed line as one
923
+ substring finds almost nothing (Dallas, 2026-10-13: 101 of 109 Marilla rows
924
+ carry the type in `str_suffix`). The page's own URL stays `propertyQueryUrl`
925
+ / `casesQueryUrl`; a `streetNumber` mapping, which no shipped city sets yet,
926
+ matches number and name on both columns instead.
927
+ - Nothing fetches. For volume against a Socrata dataset, send your own
928
+ `X-App-Token`.
929
+ - Boston's zoning record no longer names `suffolk-county`, which the atlas
930
+ does not carry (Massachusetts is statewide); its county name stays.
931
+
932
+ ## 0.6.30 — 2026-10-12
933
+
934
+ 0.6.30: endpoint capability keys supportsPagination and maxRecordCount,
935
+ offersPagination() reader; no data change. Builds on 0.6.29. `totals` do not move, and no
936
+ bundled endpoint carries either key yet. No breaking change.
937
+
938
+ Some ArcGIS layers refuse pagination: any `resultRecordCount` or
939
+ `resultOffset`, in any letter case, gets HTTP 200 with
940
+ `{"error":{"code":400,"message":"Pagination is not supported."}}` instead of
941
+ rows. The atlas will register such layers (Mississippi's statewide parcel
942
+ layers are next), so the record can now say so before any of them ship.
943
+
944
+ ### Added
945
+
946
+ - `EndpointRecord.supportsPagination?: false`. Only the literal `false` is
947
+ written; absent means the layer pages. The name is Esri's own
948
+ (`advancedQueryCapabilities.supportsPagination`).
949
+ - `EndpointRecord.maxRecordCount?: number`, the layer's declared per-query row
950
+ cap, present on every flagged endpoint.
951
+ - `offersPagination(endpoint)`: `false` on a flagged layer. Check it before
952
+ adding either parameter; when it is `false`, send one request without them
953
+ and narrow the query when `exceededTransferLimit` comes back.
954
+
955
+ The registry's validator enforces, on every flagged endpoint: a positive
956
+ integer `maxRecordCount`, a `sampleQuery` with neither parameter, and the same
957
+ two values on every registration of the same layer URL.
958
+
959
+ ## 0.6.29 — 2026-10-12
960
+
961
+ New counties, no API change, no breaking change. Builds on 0.6.28. `totals`
962
+ move from 303 counties / 322 endpoints to **306 counties / 324 endpoints**
963
+ (298 with a verified endpoint). All three were asked for through the site's free
964
+ tools. Jackson MS (28059), also asked for, is not in this release: the only
965
+ public layer for it has no pagination support, and the service's own probes
966
+ send a row cap, so it waits for a release that handles that.
967
+
968
+ ### New counties
969
+
970
+ - **Racine WI (55101)**, with a verified endpoint on the county's own ArcGIS
971
+ Enterprise server (`DataPortal/Parcels_NAD83_2011_STATEPLANE` FeatureServer
972
+ layer 0, 84,684 parcels, updated nightly): owner, situs, owner mailing address
973
+ and parcel id are searchable, pagination works, and the layer answers
974
+ cross-origin requests.
975
+ - **Westmoreland PA (42129)**, with a verified endpoint on the county's own
976
+ ArcGIS Enterprise server (`Upgrade/Parcels_AS400_VIEW_Upgrade` MapServer layer
977
+ 2, 191,739 parcels). The owner column `NAME` is a fixed-width 30-character
978
+ string padded with trailing spaces, so an exact `=` never matches: search with
979
+ `LIKE 'SMITH%'`, and the server rejects `TRIM()`. The service lives in a folder
980
+ named `Upgrade`, so the URL may move; the `notes` say so.
981
+ - **Casey KY (21045)**, no public endpoint: `hasPublicRest: false` and an empty
982
+ `endpoints` array. The Casey PVA publishes only through qPublic, the state
983
+ server carries no Casey parcel service, and Kentucky has no statewide parcel
984
+ layer yet. The `notes` record the evidence as of 2026-10-12 and the PVA office
985
+ to ask.
986
+
987
+ ## 0.6.28 — 2026-10-10
988
+
989
+ withholding: RESTRICTED markers and masked mailing. Data notes only, no API
990
+ change, no breaking change, no change to `totals`. Four county notes now say
991
+ how the service treats those values:
992
+
993
+ - New Hanover NC: an owner that reads "RESTRICTED" is a county redaction
994
+ marker, so that row is withheld.
995
+ - Fulton GA: 41 rows whose owner is the county's RESTRICTED marker are withheld.
996
+ - Lake IL: 24 rows whose owner (taxpayer_name) is the county's RESTRICTED marker are
997
+ withheld.
998
+ - Wood WV: 263 mailing values carry asterisks. 251 of them are fully masked
999
+ (some keep the ZIP, some carry none) and are served as no mailing address;
1000
+ the other 12 are real addresses and are served.
1001
+
1002
+ On the service side (not in this package), "NOT RESTRICTED" and "NOT
1003
+ CONFIDENTIAL" are not markers, and a mailing value counts as fully masked when
1004
+ it is only asterisks and separators, at least one asterisk, with at most a
1005
+ trailing ZIP or ZIP+4, or a run of three or more "#".
1006
+
1007
+ ## 0.6.27 — 2026-10-10
1008
+
1009
+ New counties and a New York outage fix, no API change, no breaking change. `totals` move from 294
1010
+ counties / 305 endpoints to **303 counties / 322 endpoints** (296 with a
1011
+ verified endpoint). Each new county sits on a statewide layer the atlas already
1012
+ carries, one registration per county scoped by that layer's county field:
1013
+ the NC OneMap parcel aggregation (`cntyname`, as Forsyth and Montgomery use),
1014
+ the Montana cadastral layer (`CountyName`, as Yellowstone uses) and the WVU
1015
+ parcel layer (`COUNTY`, the state's alphabetical code, as Kanawha uses).
1016
+
1017
+ ### New counties
1018
+
1019
+ - **New Hanover NC (37129)**, 103,489 parcels, owner on 99.8%.
1020
+ - **Iredell NC (37097)**, 109,254 parcels. The site city is blank on every row.
1021
+ - **Union NC (37179)**, 117,539 parcels. The site city is blank on every row.
1022
+ - **Onslow NC (37133)**, 95,297 parcels.
1023
+ - **Johnston NC (37101)**, 125,583 parcels. 68 owner rows carry an asterisk
1024
+ on a real name.
1025
+ - **Brunswick NC (37019)**, 151,753 parcels, situs on 85%. The site city is
1026
+ blank on every row.
1027
+ - **Gallatin MT (30031)**, 53,618 parcels, TaxYear 2026, situs on 83.7%.
1028
+ - **Berkeley WV (54003, code `02`)**, 62,652 parcels, situs on 98.3%.
1029
+ - **Wood WV (54107, code `54`)**, 58,263 parcels. 263 mailing addresses are
1030
+ masked with asterisks.
1031
+
1032
+ NC rows carry no reliable per-county date (`revisedate` is null and the layer
1033
+ publishes no editing info). West Virginia's layer says it was updated for Tax
1034
+ Year 2023. North Carolina goes from 9 to 15 counties, Montana from 4 to 5 and
1035
+ West Virginia from 4 to 6.
1036
+
1037
+ ### New York outage fix
1038
+
1039
+ The state parcel server (`gisservices.its.ny.gov`) has hung on every query
1040
+ since about 2026-10-09 16:00-18:00Z. Eight New York records gain the official
1041
+ NYSGIS mirror of the same layer on ArcGIS Online (`services6.arcgis.com`,
1042
+ FeatureServer/1, 2025 roll, last edited 2026-09-25, open CORS). Onondaga,
1043
+ Suffolk and Westchester list the mirror first and keep the state endpoint
1044
+ second; the five NYC boroughs list it third, behind MapPLUTO and the state
1045
+ endpoint. Counties do not change, so only the endpoint total moves (+8).
1046
+ The mirror carries no mailing address for the boroughs. A real owner name and
1047
+ street address in the Onondaga note was replaced with a generic description.
1048
+
1049
+ ## 0.6.26 — 2026-10-10
1050
+
1051
+ New counties, no API change, no breaking change. `totals` move from 286
1052
+ counties / 297 endpoints to **294 counties / 305 endpoints** (287 with a
1053
+ verified endpoint). All eight sit on the Colorado OIT statewide parcel
1054
+ composite already used by El Paso and Jefferson, one registration per county
1055
+ scoped by `countyName`. The layer is a state roll-up: each county's rows carry
1056
+ the `dateReceived` the state loaded them (all 2026-03-27 to 2026-05-19), so
1057
+ ownership can trail the county's own roll.
1058
+
1059
+ ### New counties
1060
+
1061
+ - **Arapahoe CO (08005)**, 231,438 parcels, situs on 93%, mailing on 98%.
1062
+ - **Adams CO (08001)**, 188,474 parcels, situs on 93%.
1063
+ - **Larimer CO (08069)**, 180,960 parcels, situs on 88%.
1064
+ - **Douglas CO (08035)**, 164,818 parcels, situs on 88%.
1065
+ - **Weld CO (08123)**, 163,491 parcels, situs on 81%.
1066
+ - **Boulder CO (08013)**, 144,551 parcels, situs on 99%.
1067
+ - **Mesa CO (08077)**, 78,984 parcels. 650 carry the placeholder owner
1068
+ "Confidential Owner"; those rows are withheld, not served as a name.
1069
+ - **Park CO (08093)**, 43,241 parcels, situs on 85%.
1070
+
1071
+ Pueblo (101,146 parcels) was checked and left out: the layer carries no situs
1072
+ address for it. Colorado goes from 3 to 11 counties.
1073
+
1074
+ ## 0.6.25 — 2026-10-10
1075
+
1076
+ New counties, no API change, no breaking change. `totals` move from 280
1077
+ counties / 291 endpoints to **286 counties / 297 endpoints** (279 with a
1078
+ verified endpoint). All six sit on the Arkansas GIS Office statewide parcel
1079
+ layer, one registration per county scoped by `countyfips`, and none of them
1080
+ publishes a mailing address (the layer has no mailing field).
1081
+
1082
+ ### New counties
1083
+
1084
+ - **Clay AR (05021)**, requested by a visitor through the county-request form.
1085
+ 16,324 parcels, owner populated on every row, situs on 69%.
1086
+ - **Saline AR (05125)**, 80,143 parcels, situs on 96%.
1087
+ - **Garland AR (05051)**, 76,870 parcels, situs on 99%.
1088
+ - **Faulkner AR (05045)**, 66,634 parcels, situs on 88%.
1089
+ - **Sebastian AR (05131)**, 60,283 parcels, situs on 91%.
1090
+ - **Craighead AR (05031)**, 54,896 parcels, situs on 87%.
1091
+
1092
+ Arkansas goes from 3 to 9 counties.
1093
+
1094
+ ## 0.6.24 — 2026-10-10
1095
+
1096
+ New counties, no API change, no breaking change. `totals` move from 277
1097
+ counties / 288 endpoints to **280 counties / 291 endpoints** (273 with a
1098
+ verified endpoint). Each was chosen for population and for publishing owner,
1099
+ mailing address, situs and parcel id on a public layer.
1100
+
1101
+ ### New counties
1102
+
1103
+ - **Chester PA (42029)**, with a verified endpoint on PASDA, Penn State's state
1104
+ clearinghouse (`pasda/ChesterCounty` MapServer layer 11, "Chester County
1105
+ Parcels 202610", 194,767 parcels, county-supplied). Owner, situs, owner
1106
+ mailing address and UPI are searchable, and the layer answers cross-origin
1107
+ requests. PASDA republishes monthly under a new vintage stamp, so the layer
1108
+ name (and possibly its index) rotates; the `notes` say how to re-resolve it,
1109
+ and that the older `maps.pasda.psu.edu/arcgis` copy is dead.
1110
+ - **Gaston NC (37071)**, with a verified endpoint on the county's own
1111
+ `PublicGIS/Parcels` MapServer layer 11 (118,132 parcels). The layer carries
1112
+ three owner vintages; the record lists only the current-owner `CURR_`
1113
+ columns, not the January-1 (`JAN1_`) or prior-year (`PRVYR`) ones. No CORS
1114
+ header, so it is server-side only.
1115
+ - **Clark WA (53011)**, with a verified endpoint on the county's own
1116
+ `MapsOnline/MailingLabels` FeatureServer layer 0 ("Owner Addresses", 196,284
1117
+ taxlots): owner, care-of, a complete owner mailing address, situs and the
1118
+ Assessor's property id, plus a parcel publish date (2026-10-09). The sibling
1119
+ `PropertyFinder` layer is not used because it has no date and no mailing
1120
+ city, state or ZIP. No CORS header, so it is server-side only.
1121
+
1122
+ ## 0.6.23 — 2026-10-10
1123
+
1124
+ New counties, no API change, no breaking change. Builds on 0.6.22. `totals`
1125
+ move from 273 counties / 284 endpoints to **277 counties / 288 endpoints**
1126
+ (270 with a verified endpoint). All four were asked for through the site's free
1127
+ tools.
1128
+
1129
+ ### New counties
1130
+
1131
+ - **James City VA (51095)**, with a verified endpoint on the county's own
1132
+ ArcGIS Server (`JCC/GIS_Data` FeatureServer layer 17, 36,478 parcels):
1133
+ owner, situs, owner mailing address and PIN are searchable, and the layer
1134
+ answers cross-origin requests. The county's licence text asks that a
1135
+ disclaimer be placed on all products derived from the data; the `notes`
1136
+ quote it. The county's ArcGIS Online copy of the layer requires a token and
1137
+ is not used.
1138
+ - **Hill TX (48217)**, with a verified endpoint on Hill CAD's
1139
+ `HillCADWebService` FeatureServer layer 0 (39,102 parcels), hosted on the
1140
+ appraisal district's GIS vendor's ArcGIS Online organisation. The situs is
1141
+ split over `situs_num` and `situs_street` (plus prefix and suffix); the
1142
+ record lists `situs_street` as the site-address column, so search an address
1143
+ by street name plus `situs_num`. 213 rows read `CONFIDENTIAL` as the owner
1144
+ (Texas Tax Code §25.025 withholding); the hosted API and MCP serve those rows
1145
+ with `withheld: "county-redacted"`, as described under 0.6.22. The
1146
+ vendor-hosted URL can move; the `notes` say so.
1147
+ - **Greene MO (29077)**, with a verified endpoint on the City of Springfield's
1148
+ `Maps/GisViewerGroups` MapServer layer 18 (123,884 parcels), which covers the
1149
+ whole county, not only the city. The owner mailing address is one combined
1150
+ string. The layer publishes no vintage and no terms.
1151
+ - **Georgetown SC (45043)**, parcel shapes only: the county's
1152
+ `GCGIS_OpenData` FeatureServer layer 2 (47,943 parcels) carries the TMS,
1153
+ plat and zoning and no owner, situs or mailing column, so the record carries
1154
+ an `ownerFieldNote` and `not_published` overrides for those three fields.
1155
+ The county publishes owners in a separate non-spatial table on the same
1156
+ service (layer 7, `PARCELATTRIBUTES`, keyed by TMS); the atlas registers one
1157
+ spatial layer per endpoint and does not join tables, so the `notes` describe
1158
+ the table and the two-query join rather than registering it.
1159
+
1160
+ ## 0.6.22 — 2026-10-10
1161
+
1162
+ New counties, no API change, no breaking change. Builds on 0.6.21. `totals`
1163
+ move from 257 counties / 268 endpoints to **273 counties / 284 endpoints**
1164
+ (266 with a verified endpoint). Kentucky goes from one county to 17: 13 publish
1165
+ owner names (Fayette plus 12 new) and 4 are parcel shapes only.
1166
+
1167
+ ### New counties — Kentucky, with owner names (12)
1168
+
1169
+ Boone (21015), Warren (21227), Daviess (21059), McCracken (21145), Clark
1170
+ (21049), Shelby (21211), Webster (21233), Franklin (21073), Simpson (21213),
1171
+ Henry (21103), Montgomery (21173) and Meade (21163). Each record carries the
1172
+ parcel id, the owner name, the owner mailing address, the situs address (Henry
1173
+ has none) and the county's other columns as documented `searchFields`, and a
1174
+ sample query on the owner column. Every layer answers cross-origin requests.
1175
+
1176
+ - **Withheld-address note.** Kentucky law lets judges (KRS 61.7991) and people
1177
+ in the Safe at Home program (KRS 14.304) have their home address withheld, so
1178
+ a county may blank an owner or label it `CONFIDENTIAL`. Each of the 12 records
1179
+ carries `capabilityOverrides.owner_name` and `owner_mailing_address` as
1180
+ `available` with a `statute` basis whose `note` says so, cites both statutes
1181
+ and names the address where a Safe at Home participant can send written
1182
+ notice, and says that judges and participants who send written notice have
1183
+ their parcel withheld from every UrbanKit surface. (The hosted API and MCP
1184
+ now serve a row with a county redaction marker in its owner, mailing or situs
1185
+ value as owner, mailing and situs null with `withheld: "county-redacted"`,
1186
+ keeping only the request fields and the parcel id; the package ships only the
1187
+ registry. A row withheld on written notice, `withheld: "notice-suppressed"`,
1188
+ is not charged; on `radius_owners` / `POST /api/atlas/radius`, which bill per
1189
+ row that carries an owner or mailing address, a county-redacted row is not
1190
+ charged either.) Read it with `reviewedCapability(county, "owner_name")`. These
1191
+ counties are `available`, not `restricted`.
1192
+ - **Franklin** publishes a 2019 snapshot (its newest `Year` is 2019); its
1193
+ record carries a `dataVintage` advisory that says so.
1194
+ - **Simpson**'s layer states "Data is not to be used for legal purposes"; the
1195
+ `notes` quote it. Its mailing street is `Address__1` (`Address_Li` is a
1196
+ second line used on 66 rows).
1197
+ - **Meade**'s service name ends in `_WFL1`, and that suffix is required.
1198
+ - In the Shelby, Franklin, McCracken, Webster, Henry, Montgomery, Meade and
1199
+ Simpson layers the plain `City`, `State` and `Zip` columns are the OWNER'S
1200
+ mailing city, state and ZIP, not the situs; the labels say so.
1201
+
1202
+ ### New counties — Kentucky, parcel shapes only (4)
1203
+
1204
+ Jefferson (21111, LOJIC's open-data layer), Madison (21151), Oldham (21185)
1205
+ and Henderson (21101). Their layers carry a parcel id and geometry (Henderson
1206
+ also a situs address) and no owner column, so each records `owner_name` as
1207
+ `not_published` on an `endpoint_schema` basis. Jefferson's record uses LOJIC's
1208
+ current layer rather than Louisville Metro's `New_AllParcels`, which has an
1209
+ address and a property class but was last edited in 2017.
1210
+
1211
+ ## 0.6.21 — 2026-10-08
1212
+
1213
+ Three demanded counties, no API change, no breaking change. `totals` move to
1214
+ **257 counties / 268 endpoints** (250 with a verified endpoint).
1215
+
1216
+ ### New counties
1217
+
1218
+ - **Lyon NV (32019)**, with a verified endpoint on the county's own ArcGIS
1219
+ Server (`Landbase/Land_Use_Parcel`, MapServer layer 10 `Land Use -- Parcel`,
1220
+ 35,343 parcels, owner on 99.5% of them, CORS echoes the caller's origin).
1221
+ Owner (`PANAME`), situs (`PHY_ADDR`), mailing address (`PMADD1`/`PMADD2`/
1222
+ `PMCTST`/`PZIP`) and the dashed `APN` are all on the layer. The owner column
1223
+ has no index: a `%SMITH%` search takes 13 to 15 seconds and an anchored
1224
+ `SMITH%` search about 4.5, which the `notes` record.
1225
+ - **Humboldt CA (06023)**, with a verified endpoint on the county's own ArcGIS
1226
+ Server (`Parcels/Parcels`, MapServer layer 0, 72,721 parcels). The layer is
1227
+ parcels only: it publishes the APN, situs address, zoning, general plan and
1228
+ use code, but no owner or mailing columns, and the `notes` say so.
1229
+ - **Fresno CA (06019)**, parcels only, with a verified endpoint on the county's
1230
+ own ArcGIS Server (`FC_PARCEL_SELECT`, MapServer layer 0, 442,045 rows). The
1231
+ county states it cannot provide ownership information on the internet under
1232
+ Cal. Gov. Code 6254.21 (now 7928.205), so the record carries an
1233
+ `ownerFieldNote` and `capabilityOverrides` marking `owner_name` and `owner_mailing_address`
1234
+ `restricted` (`county_cited_statute`, attributed to the County of Fresno), and
1235
+ neither column is in `searchFields`. `reviewedCapability` and
1236
+ `isReviewedUnservable` report both restricted. The APN, situs address, land
1237
+ use and assessed values are searchable.
1238
+
1239
+ ## 0.6.20 — 2026-10-08
1240
+
1241
+ New counties, no API change, no breaking change. `totals` move to **254
1242
+ counties / 265 endpoints** (247 with a verified endpoint).
1243
+
1244
+ ### New counties
1245
+
1246
+ - **Chesterfield VA (51041)**, with a verified endpoint on the county's own
1247
+ ArcGIS Online organisation (`Cadastral_ProdA`, layer 3 `ParcelsEnriched`,
1248
+ 150,624 parcels, edited daily, CORS `*`). The county's older Hub `Parcels`
1249
+ layer is deprecated and frozen at 2024-07-30; the record does not use it.
1250
+ - **Newport News VA (51700)**, an independent city, with a verified endpoint on
1251
+ the city's `Operational/Parcel` MapServer (54,444 parcels). The layer
1252
+ publishes no edit date, which the `notes` say.
1253
+ - **Bradford PA (42015)**, with a verified endpoint on PASDA's
1254
+ `BradfordCounty` MapServer layer 5 (34,935 parcels, county-supplied data).
1255
+ The owner field is truncated at 25 characters, the address columns are the
1256
+ owner's mailing address, and the layer carries no situs address; the `notes`
1257
+ say so.
1258
+ - **Montgomery NC (37123)**, with a verified endpoint on the NC OneMap
1259
+ statewide parcel layer, scoped by `scopeWhere` to
1260
+ `cntyname = 'Montgomery'` (30,283 parcels).
1261
+ - **Dutchess NY (36027)**, registered with `hasPublicRest: false`: the county's
1262
+ parcel layer carries geometry and ids only, the New York State tax parcel
1263
+ service has no Dutchess rows, and owner data is sold by the county's Real
1264
+ Property Tax Service Agency. The `notes` record the evidence and the contact.
1265
+
1266
+ ## 0.6.19 — 2026-10-06
1267
+
1268
+ New county, no API change, no breaking change. `totals` move to **249
1269
+ counties / 261 endpoints** (243 with a verified endpoint).
1270
+
1271
+ ### New county — Polk OR (41053)
1272
+
1273
+ Asked for through the site's free tools on 2026-10-04. The county's
1274
+ Hub-advertised REST path (`maps.co.polk.or.us/gis/`) has returned HTTP 500 on
1275
+ every request since at least 2026-10-05, but the same ArcGIS Server answers on
1276
+ its second web adaptor, `/arcserv2/`, and the record registers its
1277
+ `Klop_core/Acct_35` taxlot layer: 36,413 taxlots with the assessment account
1278
+ joined on, so owner name, owner mailing address, situs, Map Taxlot, ORTaxlot and
1279
+ account number are all searchable, and the layer answers cross-origin
1280
+ requests (`corsEnabled: true`), so browser clients query it directly. The
1281
+ `notes` record the outage and the adaptor it works around.
1282
+
1283
+ ## 0.6.18 — 2026-10-06
1284
+
1285
+ New county, no API change, no breaking change. `totals` move to **248
1286
+ counties / 260 endpoints** (242 with a verified endpoint).
1287
+
1288
+ ### New county — Mendocino CA (06045)
1289
+
1290
+ The County's own ArcGIS Online organisation (Mendocino County GIS) publishes a
1291
+ county-wide assessor parcel layer, `Parcels_Public_` FeatureServer layer 0:
1292
+ 62,051 parcels, data last edited 2026-06-08, CORS open. APN (8-character text),
1293
+ the 10-character APNFULL and the situs address are searchable. The layer
1294
+ publishes no owner name and no owner mailing address, so the record carries an
1295
+ `ownerFieldNote` and `not_published` capability overrides for both, and an
1296
+ owner search returns nothing for this county. The City of Ukiah's organisation
1297
+ hosts two Ukiah-area clips of the same data; neither is county-wide, so neither
1298
+ is listed. `addedAt` is 2026-10-06.
1299
+
1300
+ ## 0.6.17 — 2026-10-05
1301
+
1302
+ New county, one new optional field, no breaking change. `totals` move to **247
1303
+ counties / 259 endpoints** (241 with a verified endpoint).
1304
+
1305
+ ### New county — Schenectady NY (36093)
1306
+
1307
+ Asked for through the site's free tools on 2026-10-04 and not in the NYS Tax
1308
+ Parcels Public statewide layer (a count on `COUNTY_NAME='Schenectady'` returns
1309
+ 0 where the Albany control returns 112,804). The record registers the public
1310
+ hosted layer published by Spatial Data Logic, the vendor behind the county's
1311
+ property system: 59,235 parcels, owner names published, parcel id (tax-map
1312
+ print key), situs and owner mailing address. It carries a `dataVintage`
1313
+ advisory, because the layer's data was last edited on 2023-09-28.
1314
+
1315
+ ### New field — `addedAt`
1316
+
1317
+ A county may now carry `addedAt` (`YYYY-MM-DD`), the date it entered the
1318
+ atlas. It is optional (counties added before this release carry none) and,
1319
+ unlike `lastVerified`, it never moves. Schenectady is the first record to carry
1320
+ it.
1321
+
1322
+ ### Re-checked, still no public REST — Iron MI (26071), Garfield OK (40047)
1323
+
1324
+ Both were asked for and both stay `hasPublicRest: false`; their `notes` now
1325
+ record the 2026-10-05 re-check (Michigan's state ArcGIS server lists no parcel
1326
+ layer; Garfield's Spatialest GeoServer still has WFS disabled, and Oklahoma has
1327
+ no statewide parcel service).
1328
+
1329
+ ## 0.6.16 — 2026-10-04
1330
+
1331
+ Data refresh. No API change. `totals` unchanged at **246 counties / 258
1332
+ endpoints**. Five New Jersey county records on the two NJOGIS statewide
1333
+ services now state what those services serve. All 67 Florida registrations
1334
+ on the FDOR statewide cadastral replace a `sampleQuery` the layer rejects
1335
+ with one scoped to the county. The site's downloadable directory now scopes its sample queries,
1336
+ which is not part of this package.
1337
+
1338
+ ### Data refresh — Bergen NJ lists the mailing-address columns its shared layer serves
1339
+
1340
+ - **New Jersey:** Bergen County (FIPS 34003) sits on the NJ Office of GIS
1341
+ statewide composite, `Parcels_Composite_NJ_WM` FeatureServer layer 0, scoped
1342
+ `COUNTY='BERGEN'`. Essex County uses the same layer, and Essex's record
1343
+ already listed the mailing address on the tax record. Bergen's did not, so
1344
+ it understated what the layer returns. Measured live 2026-10-04, with
1345
+ Essex as the control:
1346
+
1347
+ | `COUNTY=` scope, test | BERGEN | ESSEX |
1348
+ |---|---|---|
1349
+ | all rows | 281,646 | 178,089 |
1350
+ | `ST_ADDRESS<>''` | 281,646 | 178,089 |
1351
+ | `CITY_STATE<>''` | 281,646 | 178,089 |
1352
+ | `ZIP_CODE<>''` | 281,612 | 178,088 |
1353
+ | `OWNER_NAME<>''` | 0 | 0 |
1354
+ | `PROP_LOC<>''` | 281,646 | 178,089 |
1355
+ | `CALC_ACRE>0` | 244,163 | 152,381 |
1356
+ | `YR_CONSTR>0` | 263,852 | 156,919 |
1357
+ | `BLDG_DESC<>''` | 241,129 | 121,070 |
1358
+
1359
+ `OWNER_NAME IS NOT NULL` returns every row in both counties, because the
1360
+ column holds empty strings. Test it with `<>''`.
1361
+ - **`searchFields`:** adds `ST_ADDRESS` ("Mailing Street Address",
1362
+ searchable), `CITY_STATE` ("Mailing City and State") and `ZIP_CODE`
1363
+ ("Mailing ZIP Code"), worded exactly as Essex words them. Adds `CALC_ACRE`
1364
+ ("Calculated Acreage"). **Removes `OWNER_NAME`**, as Essex does: it is
1365
+ blank on every row statewide, and the owner-field classifier takes the first
1366
+ owner-looking column, so listing it pointed owner lookups at a blank
1367
+ column. The column's absence of data stays on record through the existing
1368
+ `ownerFieldNote` and the reviewed `capabilityOverrides.owner_name`
1369
+ (`restricted`, Daniel's Law, quoted from the service description). Both are
1370
+ unchanged apart from the row count in the note.
1371
+ - `sampleQuery` now returns the mailing columns. `corsEnabled` `null` →
1372
+ `true` (`Access-Control-Allow-Origin: *`). `lastVerified` 2026-05-15 →
1373
+ 2026-10-04. `licenseUrl` and `contact.url` were `https://www.njogis.com/`,
1374
+ which no longer resolves (NXDOMAIN). They now point to the layer's ArcGIS
1375
+ item and to NJGIN, as Essex's do. `notes` rewritten to the measured facts.
1376
+ The old notes claimed "20,553+ parcels", and the scope holds 281,646.
1377
+ - **What a consumer sees:** the site's derived capability for
1378
+ `owner_mailing_address` on Bergen moves from `not_published` to
1379
+ `available`, the same as Essex. Owner-name search stays unavailable.
1380
+ - **Essex** (FIPS 34013) now lists `YR_CONSTR` ("Year of Construction") and
1381
+ `BLDG_DESC` ("Building Description") with Bergen's labels. Both are
1382
+ populated for Essex: 156,919 and 121,070 of 178,089 rows. With that change
1383
+ Bergen and Essex list the same composite columns. `lastVerified` is now
1384
+ 2026-10-04 and `notes` carry the two counts.
1385
+ - **Ocean (34029), Camden (34007) and Morris (34027)** stay on their own
1386
+ service, `maps.nj.gov` `Framework/Cadastral/MapServer/0`. It publishes the
1387
+ same 3,481,240 statewide rows as the composite, and its Essex counts match
1388
+ the composite's exactly. Measured on that service 2026-10-04:
1389
+
1390
+ | `COUNTY=` scope, test | OCEAN | CAMDEN | MORRIS | ESSEX (control) |
1391
+ |---|---|---|---|---|
1392
+ | all rows | 299,346 | 191,292 | 177,533 | 178,089 |
1393
+ | `ST_ADDRESS<>''` | 299,346 | 191,292 | 177,533 | 178,089 |
1394
+ | `CITY_STATE<>''` | 299,346 | 191,292 | 177,531 | 178,089 |
1395
+ | `ZIP_CODE<>''` | 299,327 | 191,292 | 177,514 | 178,088 |
1396
+ | `PROP_LOC<>''` | 299,346 | 191,292 | 177,531 | 178,089 |
1397
+ | `OWNER_NAME<>''` | 0 | 0 | 0 | 0 |
1398
+ | `CALC_ACRE>0` | 260,309 | 169,212 | 166,312 | 152,381 |
1399
+ | `YR_CONSTR>0` | 258,549 | 162,106 | 156,490 | 156,919 |
1400
+ | `BLDG_DESC<>''` | 260,928 | 153,496 | 146,953 | 121,070 |
1401
+
1402
+ For each of the three:
1403
+ - **`OWNER_NAME` is removed** from `searchFields`. It was listed as
1404
+ searchable, and Ocean's notes called it "public, queryable".
1405
+ - **Fields added:** `ST_ADDRESS`, `CITY_STATE` and `ZIP_CODE` with Essex's
1406
+ labels. `CALC_ACRE`, `YR_CONSTR` and `BLDG_DESC` are listed with Bergen's
1407
+ labels.
1408
+ - **`PAMS_PIN` relabelled** "PAMS Parcel Identifier (MUN_BLOCK_LOT)". The old
1409
+ label, "PAMS Property Identification Number", did not match the site's
1410
+ parcel-id reader, so `apn` derived as `not_published`. It now derives as
1411
+ `available`, and so does `owner_mailing_address`.
1412
+ - **`capabilityOverrides.owner_name` stays `not_published`**, with its
1413
+ `endpoint_schema` basis. This service states no reason for the blank
1414
+ column. The Daniel's Law attribution comes from the composite service's
1415
+ description, so `ownerFieldNote` and `notes` quote it as that service's
1416
+ statement. They do not record it as a restriction this endpoint cites.
1417
+ - **Other changes:** `notes` and `ownerFieldNote` are rewritten from the
1418
+ measured counts. Camden's layer carries 36 municipality names. Its old
1419
+ notes said 37 municipalities. Unverified population figures and clerk links
1420
+ are dropped, including oceanrecorder.com, which no longer resolves.
1421
+ - `corsEnabled` `null` → `true`, because the service reflects the request
1422
+ origin. `sampleQuery` is now a `PROP_LOC` search that returns the mailing
1423
+ columns. The old one searched the blank owner column and returned nothing.
1424
+ `lastVerified` 2026-05-15 → 2026-10-04.
1425
+ - **Bergen and Essex** `capabilityOverrides.owner_name.basis.sourceUrl` and
1426
+ `ownerFieldNote.source.url` move from `https://www.nj.gov/njgin/`, which
1427
+ does not carry the Daniel's Law statement, to the composite's ArcGIS item,
1428
+ `https://www.arcgis.com/home/item.html?id=533599bbfbaa4748bf39faf1375a8a9c`.
1429
+ Its summary reads "The OWNER_NAME field is redacted per Daniel's Law for
1430
+ public web access." Essex's `notes` and `ownerFieldNote` are dated
1431
+ 2026-10-04, matching its `lastVerified`. The counts are unchanged: 178,089
1432
+ rows, 0 with an owner name, 3,481,240 statewide, all re-measured 2026-10-04.
1433
+
1434
+ ### Data fix — Florida FDOR sample queries use a capped scoped `where`
1435
+
1436
+ - **Florida:** 57 county records use the Florida Department of Revenue
1437
+ statewide cadastral (`Florida_Statewide_Cadastral` FeatureServer layer 0)
1438
+ as their first endpoint. Their `sampleQuery` was a point+distance query.
1439
+ On 2026-10-04 the layer rejects every distance-buffered query with HTTP 200
1440
+ and error 400 "24204: The spatial reference identifier (SRID) is not
1441
+ valid", whether the point is sent in wkid 4326 or 3857, so the sample
1442
+ returned nothing. Each record's `sampleQuery` is now
1443
+ `where=(CO_NO=<n>)&outFields=*&resultRecordCount=1`. That is the record's
1444
+ own `scopeWhere`, composed the way the site's directory composes it. Fired
1445
+ live 2026-10-04: all 57 answered HTTP 200 with one row carrying that
1446
+ county's own `CO_NO`, median about 1 second, slowest under 10 seconds.
1447
+ Without the cap the layer returns its 2,000-row maximum. No `scopeWhere`,
1448
+ URL, field list or note changed.
1449
+ - The 10 Florida records that register FDOR as a **second** endpoint, after
1450
+ their own county layer, get the same sample. Fired live 2026-10-04:
1451
+ Duval, Manatee, Sarasota, Pinellas and Pasco answered one row with their
1452
+ own `CO_NO` in about half a second. Broward answered after 55 seconds cold
1453
+ and in 50 ms once warm. Miami-Dade, Hillsborough, Palm Beach and Lee did
1454
+ not answer it: about 55 seconds each, then HTTP 200 with error 400 "Unable
1455
+ to perform query", twice each. On those large counties the layer also
1456
+ fails a cap of 101 or 501, except Hillsborough, which answers both. That
1457
+ matches the slow-scan limit the Florida notes describe, because the layer
1458
+ does not index `CO_NO`. The old point+distance sample failed on all ten.
1459
+
1460
+ ## 0.6.15 — 2026-10-03
1461
+
1462
+ SDK change, no data change. `totals` stay at **246 counties / 258
1463
+ endpoints**. This release lets an agent tell which layers take a typed search,
1464
+ and fixes the deep-link helper, which sent people to a search that cannot run
1465
+ (GitHub #905). One existing output changes; see "Changed".
1466
+
1467
+ 68 endpoints carry `attributeSearch: "unsupported"`: the 67 Florida counties on
1468
+ the shared FDOR cadastral layer, whose county column is not indexed, and
1469
+ Orleans Parish LA, whose server never finishes a table scan. Those layers
1470
+ answer spatial queries and return owner, address and parcel fields on every
1471
+ hit. They do not answer a `where` clause. Until now the package exposed the
1472
+ flag but no reader for it, and `buildParcelLookupDeepLink` ignored it.
1473
+
1474
+ ### Added
1475
+
1476
+ - **`offersAttributeSearch(endpoint)`**: false when the layer refuses a
1477
+ `where` clause. Query such a layer by geometry, as its `sampleQuery` shows.
1478
+ It gives the same answer as the reader urbankitstudio.com uses, on every
1479
+ bundled endpoint; a monorepo test holds the two equal.
1480
+ - **`countyOffersAttributeSearch(county)`**: true when ANY of the county's
1481
+ layers takes a `where` clause. Miami-Dade answers true through its own layer
1482
+ although its FDOR layer is flagged.
1483
+ - **`buildParcelLookupDeepLink(endpoint, base?, county?)`**: an optional third
1484
+ argument, the `CountyRecord` (or any `{ stateSlug, countySlug, endpoints? }`,
1485
+ exported as the `DeepLinkCounty` type). When given, the county's PRIMARY
1486
+ layer (`endpoints[0]`) links to the county route
1487
+ `/tools/parcel-lookup/<state>/<county>` if it is flagged or shared
1488
+ (`scopeWhere` set). Only that route knows the county, so only it can scope a
1489
+ shared layer and show a flagged one as lookup by location. It always loads
1490
+ the primary layer, so any other endpoint keeps the generic `?endpoint=` link
1491
+ to itself: Broward's flagged FDOR layer sits behind the county's own layer,
1492
+ as it does in nine other Florida counties. Passed without `endpoints`, the
1493
+ county is taken as vouching that the endpoint is its primary.
1494
+ - **`canonicalLayerUrl(url)`**: one key per layer, used for that comparison.
1495
+ An http(s) URL keys on host and path, with the scheme, a default port, the
1496
+ query string, the fragment, a trailing `/query`, repeated slashes, dot
1497
+ segments, trailing slashes and letter case folded away. urbankitstudio.com
1498
+ keys its own registry lookup the same way.
1499
+
1500
+ ### Changed — existing output, as a bug fix
1501
+
1502
+ - `buildParcelLookupDeepLink` no longer adds `fieldHint` for a flagged layer,
1503
+ with or without the new argument. This changes what an existing call
1504
+ returns: Orleans Parish lost `fieldHint=OWNERNME1`, which pre-selected an
1505
+ owner search that never answers. The FDOR layers had no `fieldHint` before.
1506
+ Calls on unflagged layers return exactly what they returned in 0.6.14. It
1507
+ ships as a patch because the old link opened a search that never answers,
1508
+ and before 1.0 a bug fix that changes a returned string is still a patch.
1509
+ - The `attributeSearch` doc comment counted 67 flagged registrations. It now
1510
+ says 68 and names Orleans Parish.
1511
+ - The README's deep-link example showed `fieldHint=TaxName` for Kane County.
1512
+ The helper never produced that, because `TaxName` does not match its owner
1513
+ rule. The example now shows the real output.
1514
+
1515
+ ## 0.6.14 — 2026-10-03
1516
+
1517
+ Data refresh. No API change. `totals` move from **241 counties / 253
1518
+ endpoints** to **246 counties / 258 endpoints**; 240 of the 246 now carry a
1519
+ verified endpoint (was 235 of 241). Five endpoints are added: four new counties
1520
+ and Clackamas OR, which was already on record without one. Garfield OK is
1521
+ added without an endpoint. Only Sullivan TN adds an owner-name search. The
1522
+ other four endpoints answer by parcel number and, except Tulare, by address,
1523
+ and each records why owner data is absent in `capabilityOverrides` and
1524
+ `ownerFieldNote`.
1525
+
1526
+ ### Data refresh — five endpoints, one county without one
1527
+
1528
+ - **New Jersey:** Essex County (FIPS 34013, Newark), on the NJ Office of GIS
1529
+ statewide Parcels and MOD-IV composite (`Parcels_Composite_NJ_WM`, layer 0),
1530
+ scoped `COUNTY='ESSEX'` (178,089 of 3,481,240 rows). Searchable by PAMS
1531
+ parcel identifier (`PAMS_PIN`), property location (`PROP_LOC`),
1532
+ municipality (`MUN_NAME`) and the mailing street on the tax record
1533
+ (`ST_ADDRESS`). **Not searchable by owner:** `OWNER_NAME` is an empty string
1534
+ on every row statewide, and the NJ Office of GIS states in the service
1535
+ description that it is redacted per Daniel's Law for public web access, so
1536
+ the column is not listed and `owner_name` is `restricted`, attributed to
1537
+ NJOGIS, as for Bergen. Licence: public. Because the composite is now shared,
1538
+ Bergen's registration gains the matching `scopeWhere` `COUNTY='BERGEN'`.
1539
+ - **Oregon:** Clackamas County (FIPS 41005, Oregon City), already on record
1540
+ without an endpoint, now carries the county's `Taxlots_CMap` FeatureServer,
1541
+ layer 0 (163,927 taxlots). Searchable by parcel number (`PARCEL_NUMBER`),
1542
+ map and taxlot (`TLNO`), site address (`SITUS`) and city (`SITUS_CITY`).
1543
+ **Not searchable by owner and no mailing address:** the layer has neither
1544
+ column and the county publishes owner names on no public layer, routing owner
1545
+ questions to Assessment and Taxation; `owner_name` and
1546
+ `owner_mailing_address` are `not_published`. No scope (county layer).
1547
+ Licence: disclaimer only, recorded `unknown`.
1548
+ - **Nebraska:** Phelps County (FIPS 31137, Holdrege), on Nebraska's statewide
1549
+ `StatewideParcelsExternal` MapServer, layer 0, scoped `County_ID='137'`
1550
+ (7,664 of 1,154,898 rows; 59 are blank-attribute polygons, which the
1551
+ sample query skips with `Parcel_ID<>''`). Searchable by parcel ID (`Parcel_ID`,
1552
+ `State_PID`), situs address (`Situs_Address`) and legal description.
1553
+ **Not searchable by owner and no mailing address:** the statewide layer has
1554
+ neither column, and the county's gWorks vendor server exposes only a parcel
1555
+ key; `owner_name` and `owner_mailing_address` are `not_published`. CORS
1556
+ reflects the request origin. Licence: none stated, recorded `unknown`.
1557
+ - **California:** Tulare County (FIPS 06107, Visalia), on the county's
1558
+ `Parcels_(Public_View)` FeatureServer, layer 0 (166,875 parcels).
1559
+ Searchable by APN only (`APN`, nine-digit text, and `APNFormatted`) plus
1560
+ legal description. **No owner, no site address, no mailing address** in the
1561
+ layer; `owner_name`, `owner_mailing_address` and `situs_address` are
1562
+ `not_published`. `maxRecordCount` is 1000. No scope (county layer). Not the
1563
+ City of Tulare's layer at maps.tulare.ca.gov. Licence: none stated, recorded
1564
+ `unknown`.
1565
+ - **Tennessee:** Sullivan County (FIPS 47163, Blountville), on the City of
1566
+ Johnson City's regional `ParcelPublishing/TaxParcels` MapServer, layer 0,
1567
+ scoped `COUNTYNAME='Sullivan County'` (78,958 of 183,538 rows; the bare value
1568
+ `Sullivan` matches nothing). Searchable by owner name (`OWNER`), parcel ID
1569
+ (`GISLINK`, `ID`), property address (`ADDRESS`, stored street first and house
1570
+ number last) and owner mailing address (`MAILADDR`). CORS reflects the
1571
+ request origin. **Licence: `restricted`.** The item's terms say delivered
1572
+ products "are not to be resold by the purchaser for any reason and may not be
1573
+ reproduced without the written permission of the staff of the Johnson City
1574
+ GIS Division"; the record holds only the query URL and the county is served
1575
+ by live query.
1576
+ - **Oklahoma:** Garfield County (FIPS 40047, Enid), **no endpoint**
1577
+ (`hasPublicRest: false`). The assessor's Spatialest viewer is backed by a
1578
+ GeoServer that serves WMS images only with WFS disabled, and Oklahoma has no
1579
+ statewide parcel REST layer. The `notes` record the search, with controls, so
1580
+ it is not repeated.
1581
+
1582
+ All six states were already registered in `POPULATED_STATES`, so `src/*.ts`
1583
+ is unchanged and only `data/` grew. Deliberately a **patch**, for the reason
1584
+ 0.6.5 gave: adding county rows to an already-registered state is the 0.6.1
1585
+ shape, and `^0.6.0` reaches everyone already installed.
1586
+
1587
+ ## 0.6.13 — 2026-10-03
1588
+
1589
+ Data refresh. No API change. `totals` unchanged at **241 counties / 253
1590
+ endpoints**; one county moves from `unreachable` to `live`.
1591
+
1592
+ ### Data refresh — Orleans Parish LA is live again, for location queries
1593
+
1594
+ - **Louisiana:** Orleans Parish (FIPS 22071, New Orleans), `apps/property3`
1595
+ MapServer layer 15, `status` **`unreachable` → `live`**, `lastVerified`
1596
+ 2026-10-03. The layer was never down: it refuses any query that names its
1597
+ `outFields` (every named list answers HTTP 200 carrying ArcGIS error 400
1598
+ "Failed to execute query." in about 0.2 s) and never finishes a table scan
1599
+ (`1=1`, `OBJECTID>0` and `PARCELID IS NOT NULL` all ran past 40 s for a
1600
+ single row), which is the shape the old sample query and 463 consecutive
1601
+ probes used. With `outFields=*` the same layer answers a point + 30 m query
1602
+ in 0.28 s (40 features) and an 800 m envelope with geometry in 0.65 s (501
1603
+ features). The record now carries **`attributeSearch: "unsupported"`**: no
1604
+ `where`-clause search is offered, so a consumer that reads the flag offers
1605
+ lookup by location only. The owner columns (`OWNERNME1`, `OWNERNME2`) and the
1606
+ mailing address (`PSTLADDRESS`, `PSTLCITY`, `PSTLSTATE`, `PSTLZIP5`) are
1607
+ returned on every spatial hit. `sampleQuery` is an exact `PARCELID`
1608
+ equality with `outFields=*`, the one predicate shape that answers quickly.
1609
+ `corsEnabled` `null` → `true` (the header reflects the request origin).
1610
+ `notes` rewritten to the measured facts. Consumers that name their columns
1611
+ must send `*` to this layer; the site does so through its wildcard list.
1612
+ **TLS caveat, stated in `notes`:** the server sends an incomplete
1613
+ certificate chain (leaf without its Sectigo intermediate). Clients that
1614
+ fetch the missing intermediate themselves (curl on Windows, Chromium and
1615
+ Safari through the OS store) connect; Node's `fetch` fails the handshake
1616
+ with `UNABLE_TO_VERIFY_LEAF_SIGNATURE` (measured 2026-10-04). UrbanKit's
1617
+ billed enrichment and radius paths run on Vercel's Edge runtime, which
1618
+ cannot be given a certificate authority and whose handling of an incomplete
1619
+ chain is undocumented and not yet measured, so assume they cannot serve this
1620
+ parish until a keyed test on the deployed function says otherwise; today an
1621
+ Orleans row ends `county-unavailable` and is refunded, unchanged from
1622
+ before. The liveness probe pins the intermediate, so its ok describes the
1623
+ layer, not those paths. A Node consumer of this record must trust the
1624
+ Sectigo intermediate before any query runs. Tracked in urbankitstudio#906.
1625
+ **Field order:** `PARID` (the assessor's per-unit parcel id) is listed first
1626
+ and `PARCELID` (the GeoPIN, a lot id shared by every condo unit on a lot)
1627
+ second, so a consumer that keys rows by the first parcel-id field keys by
1628
+ unit: a 30 m buffer in the CBD returns 40 units on 3 lots with 39 owners.
1629
+
1630
+ ## 0.6.12 — 2026-10-03
1631
+
1632
+ Data refresh. No API change. `totals` move from **239 counties / 251
1633
+ endpoints** to **241 counties / 253 endpoints**; 235 of the 241 now carry a
1634
+ verified endpoint (was 233 of 239).
1635
+
1636
+ ### Data refresh — two counties added
1637
+
1638
+ - **Minnesota:** Mille Lacs County (FIPS 27095, Milaca), with a verified
1639
+ endpoint on the county's own `MilleLacs_Public_101` MapServer, layer 35
1640
+ (20,083 parcels with a taxpayer name). Searchable by taxpayer name
1641
+ (`TAXPAYER_NAME`, stored `LAST/FIRST`, e.g. `SMITH/BRIAN & ASHLEY`), parcel
1642
+ number (`PARCELID`), site address (`PROPERTY_ADDRESS`, `CITY_TWP_NAME`) and
1643
+ the taxpayer mailing address (`TAXPAYER_ADDRESS_1` to `_4`). The layer also
1644
+ has an `OWNER_NAME` column, but it is blank on about 97 percent of rows and
1645
+ names a different party where it is filled, so it is deliberately not listed
1646
+ and owner search uses `TAXPAYER_NAME`. CORS
1647
+ reflects the request origin. No license is stated.
1648
+ - **Missouri:** the independent City of St. Louis (FIPS 29510, no county; not
1649
+ St. Louis County, 29189), with a verified endpoint on the Assessor's
1650
+ `Assessor_Public_Parcels` MapServer, layer 11. Searchable by owner name
1651
+ (`OwnerName`, truncated to 40 characters at the source), parcel id
1652
+ (`ParcelId`), site address (`SITEADDR`, which carries runs of internal
1653
+ spaces, so search on a house number plus a street word) and the owner
1654
+ mailing address (`OwnerAddr`, `OwnerCity`). **The server sends no
1655
+ `Access-Control-Allow-Origin` header, so `corsEnabled` is `false`:** a
1656
+ browser on another origin cannot read it, and it must be queried from a
1657
+ server. No license is stated.
1658
+
1659
+ Minnesota and Missouri were already registered in `POPULATED_STATES`, so
1660
+ `src/*.ts` is unchanged and only `data/` grew. Deliberately a **patch**, for
1661
+ the reason 0.6.5 gave: adding county rows to an already-registered state is
1662
+ the 0.6.1 shape, and `^0.6.0` reaches everyone already installed.
1663
+
1664
+ ## 0.6.11 — 2026-10-03
1665
+
1666
+ Repairs three registrations the liveness probe flagged (Monmouth NJ, Prince
1667
+ George's MD, Pinal AZ), records why Greenville SC stays unreachable, fixes the
1668
+ three Arkansas scopes that returned nothing without an error, reorders
1669
+ Washoe NV's owner fields, and declares two endpoint fields the bundled data
1670
+ already carried. `totals` are unchanged: 239 counties / 251 endpoints.
1671
+
1672
+ ### Fixed: Arkansas county scopes
1673
+ - The statewide Planning_Cadastre layer now stores `countyfips` as the 3-digit
1674
+ county code. `scopeWhere` and `sampleQuery` for Pulaski, Benton and
1675
+ Washington move from '05119'/'05007'/'05143' to '119'/'007'/'143'. The old
1676
+ values matched 0 rows, so every scoped attribute search for these counties
1677
+ returned nothing. The new scopes count 180,229, 175,795 and 114,791 rows
1678
+ (2026-10-03).
1679
+
1680
+ ### Fixed: endpoints
1681
+ - Monmouth County NJ moves to the county's new ArcGIS Online layer
1682
+ (Monmouth_County_Parcels FeatureServer/0 on services1.arcgis.com). The old
1683
+ services9 GEO_MC_Parcels service is gone. Field names change: PAMSPIN,
1684
+ Owner_Name, Location, Class, Net_Value, Land_Value, Impr_Value, Sale_Date,
1685
+ Sale_Price, Year_Built, Zone. Owner mailing is now Owner_Street plus
1686
+ Owner_Csz, replacing CityStateZip, so the `mailing` role a consumer derives
1687
+ moves from a city/state/zip column to a street column. Acreage is dropped.
1688
+ Value and date fields are strings on this layer.
1689
+ - Prince George's County MD moves to the Planning Department's hosted copy
1690
+ (Property_Flattened_Py FeatureServer/0 on services1.arcgis.com), with the
1691
+ same field names. The county host answers "Layer not found" on about every
1692
+ second request. The hosted copy refreshes periodically, not live.
1693
+ - Pinal County AZ: the county renamed every field to long-form names (for
1694
+ example PARCELID to Parcel_Identification_Number, OWNERNME1 to
1695
+ First_Owner_Name). All searchFields and the sampleQuery are re-mapped. The
1696
+ layer now publishes owner mailing address (Postal_Address, Postal_City,
1697
+ Postal_State, Postal_Zip_Code_5/4), so `owner_mailing_address` for Pinal
1698
+ moves from not_published to available.
1699
+ - Greenville County SC: notes only. It stays `unreachable` because the county
1700
+ deleted its GreenvilleJS services folder.
1701
+ - Washoe County NV: LASTNAME now precedes FIRSTNAME in `searchFields`, so a
1702
+ consumer that takes the first owner column searches surnames, as the
1703
+ record's own sampleQuery does. A derived owner value for a Washoe row is
1704
+ therefore the surname alone (was the first-name column).
1705
+
1706
+ ### Added: two optional fields on `EndpointRecord`
1707
+ - `scopeWhere?: string`: the predicate ANDed into every attribute search on a
1708
+ layer shared by several counties. The bundled data already carried it, and
1709
+ the type now declares it.
1710
+ - `attributeSearch?: "unsupported"`: marks a layer that cannot serve a
1711
+ `where`-clause search. Set today on the 67 Florida FDOR registrations.
1712
+
1713
+ ## 0.6.10 — 2026-10-03
1714
+
1715
+ Data refresh. No API change. `totals` move from **235 counties / 247
1716
+ endpoints** to **239 counties / 251 endpoints**; 233 of the 239 now carry a
1717
+ verified endpoint (was 229 of 235).
1718
+
1719
+ ### Data refresh — four counties added
1720
+
1721
+ - **Georgia:** Dougherty County (FIPS 13095, Albany), with a verified
1722
+ endpoint on Albany GIS's `Parcels_Public_View` FeatureServer, covering the
1723
+ city and county jurisdictions (38,007 parcels). Searchable by owner name
1724
+ (`Name`), parcel number (`ParcelNum`), site address (`Address`) and the
1725
+ owner mailing address (`MailAddress1`, `MailCity`). String fields are
1726
+ space padded, so match with `LIKE 'SMITH%'`. The layer is in wkid 102667,
1727
+ so request `outSR=4326` for longitude and latitude. No license is stated.
1728
+ - **Alabama:** Mobile County (FIPS 01097), with a verified endpoint on the
1729
+ Mobile County Revenue Commission's `MCRC_Public_Parcels` FeatureServer
1730
+ (214,040 parcels, updated 2026-09-22). Searchable by owner name (`Name1`,
1731
+ `Name2`), parcel number (`Parcel_Number`) and site address (`PropAddr1`).
1732
+ The layer has no owner mailing-address columns. String fields are space
1733
+ padded, so match with `LIKE 'SMITH%'`. The layer is in wkid 102630, so
1734
+ request `outSR=4326` for longitude and latitude. No license is stated.
1735
+ - **Oregon:** Malheur County (FIPS 41045, Vale), with a verified endpoint on
1736
+ the county's `County_GIS_Portal_WFL1` FeatureServer (20,570 tax lots).
1737
+ Searchable by owner name (`OWNERSNAME`), map and taxlot id (`MapTaxlot`),
1738
+ site address (`SITUSADD`, `CITY`) and the owner mailing address (`ADD1`,
1739
+ `CITYSTATE`). The layer name and index are month-stamped (`TL Jun2026`,
1740
+ layer 16) and the county republishes monthly. A client that finds the
1741
+ layer gone should list the FeatureServer's layers and pick the one whose
1742
+ name starts with `TL ` and which carries `OWNERSNAME`. No license is
1743
+ stated.
1744
+ - **Washington:** Island County (FIPS 53029, Coupeville), with a verified
1745
+ endpoint on the Assessor's `Geocortex/Base` MapServer layer 0. Searchable
1746
+ by owner name (`taxpayer`), parcel id (`ParcelNo`), site address
1747
+ (`physical_addr`) and the owner mailing address (`mailing_addr1`,
1748
+ `mailing_addr_city`). Values carry trailing spaces, so trim them. CORS
1749
+ reflects the request origin rather than answering `*`. No license is
1750
+ stated.
1751
+
1752
+ Georgia, Alabama, Oregon and Washington were already registered in
1753
+ `POPULATED_STATES`, so
1754
+ `src/*.ts` is unchanged and only `data/` grew. Deliberately a **patch**, for
1755
+ the reason 0.6.5 gave: adding county rows to an already-registered state is
1756
+ the 0.6.1 shape, and `^0.6.0` reaches everyone already installed.
1757
+
1758
+ ## 0.6.9 — 2026-09-26
1759
+
1760
+ Adds the District of Columbia and fixes one helper. `totals` move from
1761
+ **234 counties / 246 endpoints** to **235 counties / 247 endpoints**; 229 of
1762
+ the 235 now carry a verified endpoint (was 228 of 234). `totals.states` moves
1763
+ from 50 to 51 because the index now lists DC beside the 50 states. It is not
1764
+ a state, so count `listStates()` entries by `slug` if you need states only.
1765
+
1766
+ ### Added: the District of Columbia
1767
+
1768
+ - `data/index.json` gains a populated `district-of-columbia` entry (abbrev
1769
+ `DC`), and `data/district-of-columbia.json` carries one record: `District
1770
+ of Columbia`, FIPS 11001, countySlug `district-of-columbia`, on DC GIS's
1771
+ Owner Polygons layer (Property_and_Land_WebMercator FeatureServer, layer 40).
1772
+ - Registered in `POPULATED_STATES`, so `findState("district-of-columbia")`,
1773
+ `findCounty("district-of-columbia", "district-of-columbia")`,
1774
+ `listCountiesByState` and `atlas.byStateSlug` resolve it. `listStates()`
1775
+ returns 51 entries.
1776
+ - The District has no counties. Its one record is the District itself, the
1777
+ county equivalent the Census Bureau lists as 11001. The layer maps land, one
1778
+ polygon per lot. A condominium building is a single lot on it, usually with
1779
+ no owner name, so an owner search finds lot owners, not unit owners.
1780
+
1781
+ ### Fixed: `countySlugFromName` knew four unit names
1782
+
1783
+ It recognised County, Parish, Borough and Census Area and appended " County"
1784
+ to every other name: `countySlugFromName("Anchorage Municipality")` returned
1785
+ `anchorage-municipality-county` while the bundled record is
1786
+ `anchorage-municipality`. The same was true of the nine Connecticut planning
1787
+ regions and Massachusetts' `Statewide` record. It now uses the site's rule,
1788
+ which also knows Municipality, City, Statewide, Planning Region and District
1789
+ of Columbia, and it trims the name first. A name ending in `City` now slugs as
1790
+ it stands, which is right for the independent cities and wrong for two
1791
+ Virginia counties the registry does not carry, James City County and Charles
1792
+ City County, whose base names end in City: the old rule gave
1793
+ `james-city-county`, this one gives `james-city`. If either joins the
1794
+ registry its stored name is the full `James City County`. If you stored slugs
1795
+ this function produced for one of those names, they change; the bundled
1796
+ `countySlug` values do not.
1797
+
1798
+ This is a **patch** for the reason 0.5.3 gave: `^0.6.0` has to reach everyone
1799
+ already installed.
1800
+
1801
+ ## 0.6.8 — 2026-09-26
1802
+
1803
+ Data refresh. No API change. `totals` move from **233 counties / 245
1804
+ endpoints / 50 states** to **234 counties / 246 endpoints / 50 states**;
1805
+ 228 of the 234 now carry a verified endpoint (was 227 of 233).
1806
+
1807
+ ### Data refresh — one county added, one record corrected
1808
+
1809
+ - **Ohio:** Mahoning County, with a verified endpoint. Its parcel layer
1810
+ answers only `outFields=*`: a query that names its fields is refused
1811
+ with "Failed to execute query." The record's `sampleQuery` already sends
1812
+ `outFields=*` and its `notes` record the quirk, so a client that builds
1813
+ its own query should request every field and read the columns it needs.
1814
+ - **Mississippi:** Hinds County's layer (MDEQ OPCGIS, HINDS_PARCELS) now
1815
+ refuses a named `outFields` list the same way, measured 2026-09-26 on
1816
+ the enrichment, radius and liveness query shapes; it answered a named
1817
+ list in May. Its `sampleQuery` now sends `outFields=*`, its `notes` say
1818
+ so, and `lastVerified` moves to 2026-09-26. The record also stops listing
1819
+ `MAILADD1`, `MCITY1`, `MSTATE1` and `MZIP1` under `searchFields`: those
1820
+ columns exist in the layer's schema but are blank on every one of its
1821
+ 114,814 parcels (`returnCountOnly` with `<field> IS NOT NULL AND <field>
1822
+ <> ''` counts 0 for each, against 114,547 for `OWNNAME` and 114,514 for
1823
+ `SITEADD`, measured 2026-09-26), so the county serves owner names and
1824
+ site addresses, not mailing addresses, and a client reading
1825
+ `searchFields` for a mailing column no longer finds one there. No count
1826
+ changes.
1827
+
1828
+ Ohio was already registered in `POPULATED_STATES`, so `src/*.ts` is
1829
+ unchanged and only `data/` grew. Deliberately a **patch**, for the reason
1830
+ 0.6.5 gave: adding county rows to an already-registered state is the 0.6.1
1831
+ shape, and `^0.6.0` reaches everyone already installed.
1832
+
1833
+ ## 0.6.7 — 2026-09-26
1834
+
1835
+ Data refresh. No API change. `totals` move from **227 counties / 241
1836
+ endpoints / 50 states** to **233 counties / 245 endpoints / 50 states**;
1837
+ 227 of the 233 now carry a verified endpoint (was 223 of 227).
1838
+
1839
+ ### Data refresh — six counties in four registered states
1840
+
1841
+ - **Michigan:** Lenawee County, with a verified endpoint, and Iron County,
1842
+ which ships with `hasPublicRest: false`. Iron's map viewer has no public
1843
+ query API, and Michigan has no statewide parcel layer to fall back on.
1844
+ - **California:** Riverside County. Its public layer carries APN, situs and
1845
+ mailing address but no owner name, and `capabilityOverrides.owner_name`
1846
+ records that as `not_published`.
1847
+ - **Maryland:** Baltimore County, the county rather than Baltimore City.
1848
+ - **Oregon:** Coos County, with a verified endpoint, and Clackamas County,
1849
+ which ships with `hasPublicRest: false` because the county does not
1850
+ publish owner names online.
1851
+
1852
+ All four states were already registered in `POPULATED_STATES`, so `src/*.ts`
1853
+ is unchanged and only `data/` grew. Deliberately a **patch**, for the reason
1854
+ 0.6.5 gave: adding county rows to an already-registered state is the 0.6.1
1855
+ shape, and `^0.6.0` reaches everyone already installed.
1856
+
1857
+ ## 0.6.6 — 2026-09-22
1858
+
1859
+ **No API change and no data change.** `totals` stay at **227 counties / 241
1860
+ endpoints / 50 states**, `src/` is untouched, and `data/` is byte-identical to
1861
+ 0.6.5. If you are on 0.6.5 you gain nothing functional by upgrading.
1862
+
1863
+ What you gain is that you can **verify where this came from**. 0.6.6 is the
1864
+ first release of this package published with a [provenance
1865
+ attestation](https://docs.npmjs.com/generating-provenance-statements): npm
1866
+ attaches one only on an OIDC publish from a **public** repository, and every
1867
+ release up to and including 0.6.5 was published from a private one, so none of
1868
+ them could have had it. From 0.6.6 the package publishes from
1869
+ [`urbankitstudio/atlas`](https://github.com/urbankitstudio/atlas), and
1870
+ `npm view @urbankitstudio/atlas` will show the signed link back to the commit
1871
+ and workflow run that built the tarball.
1872
+
1873
+ The package is still **authored** in the private monorepo — `data/` is generated
1874
+ from the parcel registry that lives there — and mirrored out. So the source of
1875
+ truth has not moved; only the thing that presses publish has.
1876
+
1877
+ Deliberately a **patch**, for the same reason 0.5.3, 0.6.1 and 0.6.5 gave: the
1878
+ README says to pin `^0.x`, and `^0.6.0` resolves to `>=0.6.0 <0.7.0`, so a patch
1879
+ reaches everyone already installed. A release that changes no API and no data is
1880
+ exactly what a patch is for, and a `0.7.0` would reach nobody without a manual
1881
+ bump — which would be the wrong shape for a change whose whole point is that
1882
+ existing users get a verifiable artifact.
1883
+
1884
+ ## 0.6.5 — 2026-09-20
1885
+
1886
+ Data refresh. No API change. `totals` move from **171 counties / 174
1887
+ endpoints / 50 states** to **227 counties / 241 endpoints / 50 states**;
1888
+ 223 of the 227 now carry a verified endpoint (was 166 of 171).
1889
+
1890
+ ### Data refresh — Florida
1891
+
1892
+ - **Florida: 11 → 67 counties, 10 → 77 endpoints.** Every Florida county is
1893
+ now registered, not just the largest ten. Florida was already a populated
1894
+ state (registered in `POPULATED_STATES` since before this release), so
1895
+ `findState`/`findCounty`/`listCountiesByState`/`atlas.byStateSlug` needed
1896
+ no source change — only `packages/atlas/data/florida.json` and
1897
+ `data/index.json` grew. `src/*.ts` is unchanged by this release.
1898
+
1899
+ Deliberately a **patch**, for the reason 0.5.3 and 0.6.1 gave: the README
1900
+ says to pin `^0.x`, and `^0.6.0` resolves to `>=0.6.0 <0.7.0` — so this
1901
+ reaches everyone already installed, and a `0.7.0` would not. Adding county
1902
+ rows to an already-registered state is exactly the 0.6.1 shape (Connecticut's
1903
+ nine planning regions, the four NYC boroughs), not the 0.4.0/0.5.0 shape
1904
+ (registering a *new* state key, which changes what an existing function
1905
+ returns for inputs it previously answered `undefined` for).
1906
+
1907
+ ## 0.6.4 — 2026-09-18
1908
+
1909
+ Data corrections. No API change. `totals` stay at **171 counties / 174
1910
+ endpoints / 50 states**; five of those counties publish no endpoint, and the
1911
+ README's coverage sentence is now checked against the data by a test, because
1912
+ the prose had been wrong twice.
1913
+
1914
+ - New York: the four boroughs that shared the id `ny-new-york` are now
1915
+ `ny-bronx`, `ny-kings`, `ny-queens` and `ny-richmond`.
1916
+ - South Carolina: Charleston County's parcels moved from `MapServer/61` to
1917
+ `MapServer/4`; re-pointed and re-verified 2026-09-17.
1918
+ - Texas: Fort Bend renamed its fields (`OwnerName` → `Owner_Name` and the
1919
+ rest); re-pointed rather than dropped, re-verified 2026-09-12.
1920
+ - California: San Diego's record now explains that every owner name comes back
1921
+ as the literal `Protected Per CA Gov Code 7928.205`.
1922
+ - Manifest: keywords gain `real-estate` and `property`; the README carries the
1923
+ Socket badge.
1924
+
1925
+ `^0.6.0` reaches everyone already installed.
1926
+
1927
+ ## 0.6.3 — 2026-09-03
1928
+
1929
+ Metadata only. No data or API change. The package author, the license
1930
+ holder and the README credit now name UrbanKit Studio instead of an
1931
+ individual. `^0.6.0` reaches everyone already installed.
1932
+
1933
+ ## 0.6.2 — 2026-09-02
1934
+
1935
+ Data corrections only. No API change. `totals` are now **171 counties / 174
1936
+ endpoints / 50 states**. A patch for the reason 0.6.1 gave: `^0.6.0` reaches
1937
+ everyone already installed.
1938
+
1939
+ ### Fixed — Oregon
1940
+
1941
+ - **Multnomah County pointed at Umatilla County's tax lots.** The layer's own
1942
+ description reads "a Umatilla County GIS layer depicting land partition and
1943
+ ownership within Umatilla County" (copyright Umatilla County GIS, 44,849 tax
1944
+ lots, and Multnomah's own layer counts 284,409). Multnomah now carries Multnomah County's
1945
+ own Tax Parcels layer on services5.arcgis.com: 56 fields, owner `NAME`,
1946
+ mailing `ADDR1`/`CITY`/`STATE`/`ZIP`, `SITUSADDR`, `DEED_DATE`, `SALE_DATE`,
1947
+ `SALE_PRICE`. Owner query verified live 2026-09-02.
1948
+ - **Lane County's path answered 404 "Service not found" on 2026-09-01** and was live again on
1949
+ 2026-09-02 (75 fields, owner query returns rows). The entry keeps its path, re-verified, and its
1950
+ notes name the county's other publication of the same parcels, `AT/AddressParcelSales/MapServer/2`
1951
+ (26 fields), as the fallback if the path drops again.
1952
+
1953
+ ### Added
1954
+
1955
+ - **Umatilla County, OR (41059)**: the tax lots layer that had been filed
1956
+ under Multnomah, now under its own county (owner `MAILING_NA`, mailing
1957
+ address, situs fields).
1958
+
1959
+ ### Tooling
1960
+
1961
+ - `scripts/audit-atlas-county-names.mjs` asks every endpoint who it says it is
1962
+ (layer and service metadata) and flags one that names another county of its
1963
+ state and not its own. Across all 174 endpoints on 2026-09-02 the only flags
1964
+ are statewide layers (Montana cadastral, Utah LIR) and New York City's
1965
+ MapPLUTO, which name several counties by design.
1966
+
1967
+ ## 0.6.1 — 2026-08-29
1968
+
1969
+ Data corrections only; no API change. `totals` are now **170 counties / 173
1970
+ endpoints / 50 states** (0.6.0 shipped 155 / 155 / 50).
1971
+
1972
+ Deliberately a **patch**, for the reason 0.5.3 gave: the README says to pin
1973
+ `^0.x`, and `^0.6.0` resolves to `>=0.6.0 <0.7.0` — so this reaches everyone
1974
+ already installed, and a `0.7.0` would not. Every change below corrects a claim
1975
+ a consumer could act on, and a correction is worth nothing if it does not
1976
+ arrive.
1977
+
1978
+ ### Fixed — claims about what a county can answer
1979
+
1980
+ - **Six counties no longer advertise owner-name search** their layers cannot
1981
+ answer (#530), and eight owner advisories that turned out to be FALSE were
1982
+ reverted (#541) — the audit predicate behind them (`COL <> ''` on an
1983
+ Oracle-backed layer) compares against NULL and is never true. The corrected
1984
+ audit re-ran across all 170 counties (#543).
1985
+ - **El Paso County CO** declares `situs_address` and `owner_mailing_address`
1986
+ as `not_published`: the columns exist and are null on every row, verified
1987
+ against a Denver control on the same layer (#543, #545).
1988
+ - **Cuyahoga County OH** re-pointed to the Fiscal Office's live layer after the
1989
+ county stopped its MyPLACE service; mailing address is now available there
1990
+ (#558). **Greenville County SC** and **Orleans Parish LA** are marked
1991
+ `unreachable`, with the measurement in `notes` (#558).
1992
+
1993
+ ### Added — counties and scoping
1994
+
1995
+ - **Connecticut** as its nine planning regions (Connecticut has no counties as
1996
+ Census geography; the retired county FIPS codes never resolved) (#538).
1997
+ - **Four NYC boroughs** and **Wayne, Macomb, Montgomery, Bucks, Chatham** (#529,
1998
+ #531).
1999
+ - Every county sharing a statewide layer now carries a `scopeWhere` (#529,
2000
+ #534, #558) — previously an owner search on a shared layer could return
2001
+ statewide matches under one county's name.
2002
+
2003
+ ### Housekeeping
2004
+
2005
+ `package-lock.json` still said 0.5.3 (never bumped at 0.6.0); the README said
2006
+ 155 counties; and this package's own test suite pinned 155 and had been red,
2007
+ unrun, since the data grew — `prepublishOnly` would have refused to publish.
2008
+ All corrected here, and a root-suite guard now keeps the pins honest.
2009
+
2010
+ ## 0.6.0 — 2026-08-13 (entry written retroactively on 2026-08-29)
2011
+
2012
+ Shipped without a changelog entry; this records it. `CountyRecord` in the
2013
+ published SDK had eleven fields while the registry it mirrors has sixteen —
2014
+ everything a human had reviewed (`ownerFieldNote`, `dataVintage`,
2015
+ `capabilityOverrides`, the related-city slugs) was in the bundle but
2016
+ unreachable through the public type. 0.6.0 exposed the reviewed capability to
2017
+ SDK consumers and documented why it does not derive (#409). Minor, because the
2018
+ public type widened.
2019
+
2020
+ ## 0.5.3 — 2026-08-05
2021
+
2022
+ Says plainly that `status` is not a live signal. No data changes; `totals` are
2023
+ unchanged (155 counties / 155 endpoints / 50 states).
2024
+
2025
+ Deliberately a **patch**, not a minor. The README tells you to pin `^0.x`, and
2026
+ for a `0.x` release `^0.5.2` resolves to `>=0.5.2 <0.6.0` — so a `0.6.0` would
2027
+ not reach anyone already installed. The point of this release is to correct a
2028
+ misleading claim, which is worth nothing if it does not arrive.
2029
+
2030
+ ### Fixed — a claim that read as a live signal
2031
+
2032
+ `status` and `lastVerified` describe what was true **when the version you
2033
+ installed was published**. This package ships static JSON and makes no network
2034
+ calls, so it cannot know anything more recent. `status: "live"` therefore means
2035
+ "a check passed before publish", not "this endpoint is up right now".
2036
+
2037
+ The gap is real and was measured: on 2026-08-04, Will County IL was being served
2038
+ as `status: "live"`, `lastVerified: "2026-06-27"` while the liveness probe had it
2039
+ **down with 4 consecutive failures**.
2040
+
2041
+ ### Added
2042
+
2043
+ - **`LIVE_STATUS_URL`** — exported as a runtime value so nobody hardcodes the
2044
+ endpoint. Points at `https://urbankitstudio.com/api/atlas/status`: keyless,
2045
+ refreshed every two hours.
2046
+ - Type-level documentation on `EndpointStatus`, `status` and `lastVerified`
2047
+ explaining what each actually asserts — visible in your editor on hover.
2048
+ - README section covering the two answers people get wrong: **`found: false`**
2049
+ means no observation exists (the county may be untracked, or never probed), and
2050
+ a **503** means the health store was unreachable. Neither says the endpoint is
2051
+ up. Treat both as unknown.
2052
+
2053
+ ## 0.5.2 — 2026-08-05
2054
+
2055
+ Data content patch. Published without a changelog entry at the time; recorded
2056
+ here after the fact.
2057
+
2058
+ ### Data refresh
2059
+
2060
+ - Resynced `packages/atlas/data/` to `public/data/atlas/`, clearing **all
2061
+ remaining dead-host endpoint references**. Montana moved to
2062
+ `gisservice.mt.gov/.../MapServer/1`. Verified against the published tarball:
2063
+ 58 files, 51 state JSONs, zero dead-host references.
2064
+
2065
+ ## 0.5.1 — 2026-07-20
2066
+
2067
+ Data content patch. No new counties, no API changes — `totals` are unchanged
2068
+ (155 counties / 155 endpoints / 50 states).
2069
+
2070
+ ### Data refresh
2071
+
2072
+ - Resynced `packages/atlas/data/` to `public/data/atlas/`, which had drifted
2073
+ ahead since the 0.5.0 sync (2026-06-28):
2074
+ - **King County, WA** — `notes` updated to replace the dead-end reference to
2075
+ `tise-fzs9` (a Metro vanpool-ridership dataset, not parcels) with the two
2076
+ live Socrata tables that carry King County's PIN format: Property Legal
2077
+ Descriptions (`4854-i48r`, filter by `parcel_number`) and Real Property
2078
+ Tax Receivables (`dkna-i698`, filter by `account_number`). Resolve an
2079
+ address to a PIN via eReal Property or Parcel Viewer first, then query
2080
+ either table with that PIN.
2081
+ - **Will County, IL** — added `ownerFieldNote`, a `CountyAdvisory` pointing
2082
+ consumers to the Will County Supervisor of Assessments' property search
2083
+ portal, since the public parcel REST layer exposes only the PIN (no
2084
+ owner-name or address field).
2085
+ - `data/index.json`: content (totals, `lastUpdated`, `states`) synced
2086
+ verbatim from `public/data/atlas/index.json`; the `version` field is kept
2087
+ on the SDK's own release lineage (0.5.0 → 0.5.1) rather than copied from
2088
+ the main site's independent content-version field (which reads `0.3.0`
2089
+ and tracks separately) — consistent with how this field has been set at
2090
+ every prior sync (0.1.0 → 0.5.0).
2091
+
2092
+ ### Breaking changes
2093
+
2094
+ None. Data-content correction only. Pin `^0.x` to track patches.
2095
+
2096
+ ## 0.5.0 — 2026-06-28
2097
+
2098
+ Data resync to the live atlas (**128 → 155 counties**). All 50 US states are
2099
+ now populated.
2100
+
2101
+ ### Data refresh
2102
+
2103
+ - Synced `packages/atlas/data/` to `public/data/atlas/` (+27 counties). Eleven
2104
+ states gained their first verified endpoints: Arkansas, Delaware, Hawaii,
2105
+ Maryland, Montana, New Mexico, North Dakota, Oklahoma, South Dakota,
2106
+ West Virginia, and Wyoming. Populated coverage went from 39 to all 50 states;
2107
+ several existing states also grew. `data/index.json` totals now read 155
2108
+ counties / 155 endpoints across 50 states.
2109
+ - Registered the eleven new states in the SDK's `POPULATED_STATES` map so
2110
+ `findState` / `findCounty` / `listCountiesByState` / `atlas.byStateSlug`
2111
+ resolve them. The smoke test's full-coverage guard (every `populated: true`
2112
+ state resolves through `byStateSlug`) now pins all 50.
2113
+
2114
+ ## 0.4.0 — 2026-05-31
2115
+
2116
+ Data resync to the live atlas (**117 → 128 counties**) plus a new FIPS lookup
2117
+ and a registry fix that makes 14 already-bundled states actually reachable.
2118
+
2119
+ ### New API
2120
+
2121
+ - `findCountyByFips(fips)` — look up a county by its 5-digit county FIPS code
2122
+ (state FIPS + county FIPS, e.g. `"17089"` → Kane County, IL). Scans all
2123
+ populated states. Non-5-digit / unknown / non-string input returns
2124
+ `undefined`; it never throws.
2125
+
2126
+ ### Bug fixes
2127
+
2128
+ - **14 bundled states were unreachable.** v0.3.0 shipped data files for
2129
+ Alabama, Alaska, Connecticut, Idaho, Kentucky, Louisiana, Maine,
2130
+ Massachusetts, Mississippi, New Hampshire, New Jersey, Rhode Island,
2131
+ South Carolina, and Vermont, but they were never added to the SDK's
2132
+ `POPULATED_STATES` map — so `findState` / `findCounty` / `listCountiesByState`
2133
+ / `atlas.byStateSlug` returned `undefined`/empty for all of them despite the
2134
+ JSON being present in the package. All 39 populated states are now registered.
2135
+ A new smoke test pins that every `populated: true` state in the index resolves
2136
+ through `byStateSlug` so this can't regress.
2137
+
2138
+ ### Data refresh
2139
+
2140
+ - Synced `packages/atlas/data/` to `public/data/atlas/` (+11 counties). Notable
2141
+ growth since v0.3.0: Florida 7 → 11, Texas 5 → 10, Ohio 5 → 7. Package
2142
+ `data/index.json` totals now read 128 counties / 128 endpoints across 50
2143
+ states (39 populated).
2144
+
2145
+ ### Breaking changes
2146
+
2147
+ None. New function + additive data only. Pin `^0.x` to track minor adds.
2148
+
2149
+ ## 0.3.0 — 2026-05-16 (consolidated overnight expansion)
2150
+
2151
+ Atlas grew **36 → 117 counties** across **17 → 39 populated states**. Five
2152
+ parallel expansion batches landed PRs #34, #35, #36, #37 + earlier #28.
2153
+
2154
+ ### New counties (81 added since v0.1.0)
2155
+
2156
+ **Batch 1 (PR #28)** — 10 counties: Orange CA, San Bernardino CA, San Diego CA,
2157
+ El Paso CO, Jefferson CO, Marion IN, Mecklenburg NC, Multnomah OR, Davidson
2158
+ TN (Nashville), Salt Lake UT. Added IN/OR/TN/UT as new states.
2159
+
2160
+ **Batch 2 (PR #34)** — 10 Tier-2 metros: Erie NY (Buffalo), Monroe NY
2161
+ (Rochester), Onondaga NY (Syracuse), Hamilton OH (Cincinnati), Ada ID (Boise),
2162
+ Dane WI (Madison), Milwaukee WI, Bergen NJ, Hartford CT (statewide CT CAMA),
2163
+ Anchorage AK (statewide AK). Added ID/WI/CT/AK as new states.
2164
+
2165
+ **Batch 3 (PR #36)** — 15 Pacific NW + West: Lane OR (Eugene), Marion OR
2166
+ (Salem), Spokane WA, Snohomish WA, Thurston WA (Olympia), Davis UT, Utah UT
2167
+ (Provo), Weber UT (Ogden), Washoe NV (Reno), Pinal AZ, Santa Clara CA,
2168
+ Alameda CA, Contra Costa CA, Sacramento CA, Ventura CA. (CA-suppressed-owner
2169
+ counties indexed for completeness; OR/WA/UT/AZ varies by state law.)
2170
+
2171
+ **Batch 4 (PR #32)** — 11 Northeast entries: Suffolk MA + Middlesex MA +
2172
+ Worcester MA + Norfolk MA + Essex MA + Plymouth MA (via MassGIS L3 statewide
2173
+ composite, one entry), Chittenden VT (Burlington), Cumberland ME + York ME,
2174
+ Monmouth NJ + Ocean NJ + Camden NJ + Morris NJ (NJOGIS statewide + Monmouth
2175
+ direct), Hillsborough NH + Rockingham NH (no public REST documented),
2176
+ Providence RI (no public REST documented). Added MA/VT/ME/RI/NH as new states.
2177
+
2178
+ **Batch 5 (PR #37)** — 20 Southeast counties: Duval FL (Jacksonville),
2179
+ Manatee FL, Sarasota FL, Cobb GA (Atlanta NW), DeKalb GA (Atlanta east),
2180
+ Forsyth NC, Guilford NC, Cumberland NC, Buncombe NC, Durham NC, Greenville SC,
2181
+ Charleston SC, Jefferson AL (Birmingham), Shelby TN (Memphis), Hamilton TN
2182
+ (Chattanooga), Hinds MS (Jackson), DeSoto MS, East Baton Rouge Parish LA,
2183
+ Orleans Parish LA (New Orleans), Fayette KY (Lexington). Added AL/SC/MS/LA/KY
2184
+ as new states. Also: corrected Broward FL endpoint from FDOR statewide
2185
+ (partitioned, unreliable cross-county filter) to BCPA's own server.
2186
+
2187
+ **Batch 6 (PR #35)** — 15 Midwest counties: Kent MI (Grand Rapids), Oakland MI
2188
+ (Detroit NW), Ramsey MN (St. Paul), Waukesha WI, Brown WI (Green Bay),
2189
+ Hamilton IN (Indy suburbs), Allen IN (Fort Wayne), Montgomery OH (Dayton),
2190
+ Stark OH (Canton), Johnson IA (Iowa City), St. Louis County MO, Shawnee KS
2191
+ (Topeka), Douglas NE (Omaha), Lancaster NE (Lincoln), Sarpy NE. Added
2192
+ MI/IA/MO/KS/NE as new states.
2193
+
2194
+ ### New states (22 added since v0.1.0)
2195
+
2196
+ Alabama, Alaska, Connecticut, Idaho, Indiana, Iowa, Kansas, Kentucky,
2197
+ Louisiana, Maine, Massachusetts, Michigan, Mississippi, Missouri, Nebraska,
2198
+ New Hampshire, New Jersey, Oregon, Rhode Island, South Carolina, Tennessee,
2199
+ Utah, Vermont, Wisconsin.
2200
+
2201
+ ### Engineering / hygiene fixes
2202
+
2203
+ - **Data sync**: 5 counties (Pima AZ, Suffolk NY, Westchester NY, Dallas TX,
2204
+ Bexar TX) were in `public/data/atlas/` but missing from `packages/atlas/data/`
2205
+ in v0.1.0 and v0.2.0. Synced — these counties now ship in the npm package.
2206
+ - **State registration**: indiana, oregon, tennessee, utah, maine, massachusetts,
2207
+ new-hampshire, new-jersey, rhode-island, vermont, alaska, connecticut, idaho,
2208
+ wisconsin, michigan, iowa, missouri, kansas, nebraska, alabama, south-carolina,
2209
+ mississippi, louisiana, kentucky added to the SDK's POPULATED_STATES map.
2210
+ - **Auto-prerender**: `scripts/sync-prerender-routes.mjs` (in the main app, not
2211
+ the SDK) now auto-generates the reactSnap include list from atlas data — no
2212
+ more manual route-list updates per atlas expansion.
2213
+
2214
+ ### Known non-issues
2215
+
2216
+ - Several counties don't expose owner-name on their public REST layer due to
2217
+ state law (CA Gov Code 7928.205, WA RCW 42.56.070(8)) or county policy.
2218
+ These are still indexed for completeness — the atlas marks each county's
2219
+ searchable fields, so downstream consumers can filter for owner-name
2220
+ availability.
2221
+ - NH Hillsborough + NH Rockingham + RI Providence: no public REST documented.
2222
+ Indexed with empty endpoints so atlas consumers can show "Not Yet Indexed"
2223
+ banners + invite contributions.
2224
+
2225
+ ### Breaking changes
2226
+
2227
+ None. Backward-compatible additions only. Pin `^0.x` to track minor adds.
2228
+
2229
+ ## 0.2.0 — 2026-05-15
2230
+
2231
+ (Superseded by 0.3.0 — see above for the consolidated changelog covering all
2232
+ expansion batches that landed between v0.2.0 and v0.3.0 publish.)
2233
+
2234
+ - Added 10 new counties: Orange CA, San Bernardino CA, San Diego CA, El Paso
2235
+ CO, Jefferson CO, Marion IN, Mecklenburg NC, Multnomah OR, Davidson TN, Salt
2236
+ Lake UT
2237
+ - Added 4 new states: Indiana, Oregon, Tennessee, Utah
2238
+ - Atlas: 46 counties across 21 states
2239
+
2240
+ ## 0.1.0 — 2026-05-15
2241
+
2242
+ - Initial public release
2243
+ - 36 counties across 17 states