add-county
/add-county
Pisze plik county.yaml dla powiatu spoza pakietu: mapuje dane działek, kody zabudowy i burze z regionu, z biurem NWS.
- Autor
- qdevery ↗
- Repozytorium
- canopy-based-leads ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 07.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 07.11.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Pisze plik county.yaml dla powiatu spoza pakietu: mapuje dane działek, kody zabudowy i burze z regionu, z biurem NWS.
Wyzwalacze z opisu
Oryginalny opis (EN)
Write a canopy-based-leads county.yaml for a county that isn't bundled - find the appraisal district's bulk parcel data, map its columns and local land-use / building codes to standard tags, pick the NWS office and the region's defining storms. Use when someone wants a territory list for a new county.
Jak działa
- 1. Find the data (public pages only)
- 2. Inspect before mapping
- 3. Write county.yaml
- 4. Test
Podgląd SKILL.md
--- name: add-county description: Write a canopy-based-leads county.yaml for a county that isn't bundled - find the appraisal district's bulk parcel data, map its columns and local land-use / building codes to standard tags, pick the NWS office and the region's defining storms. Use when someone wants a territory list for a new county. --- # Add a county A county config tells the pipeline where the parcel data is, what its columns are called, and what its local codes mean. Start from the bundled examples in `src/canopy_leads/configs/counties/` (repository root = two folders above this skill): `example_county.yaml` uses the `generic` adapter, which fits most counties; `harris_tx.yaml` shows a district with several linked tables (its own adapter). ## 1. Find the data (public pages only) - Search for "<county> appraisal district bulk data", "<county> assessor parcel download" or "<county> GIS open data parcels". In Texas every county has an appraisal district (CAD); elsewhere it's the assessor or the county GIS office. - You need: owner name, mailing name / address (the "C/O" line names management firms), situs address, acreage, and land-use / property-class codes, plus parcel polygons. - Read the download page's terms. If data is behind a login, a paid order form or a CAPTCHA, stop and tell the user: they can request it themselves (often a free public-records request). - Many states also publish licensed-facility lists (assisted living, nursing homes) with coordinates; they confirm senior living and name the operator. ## 2. Inspect before mapping Download a sample (or the file) into `runs/<county>/raw/` and look at it with pandas / geopandas: column names, a few rows, the distinct values and counts of each code column. Find the code descriptions (usually a lookup table or PDF on the same site). ## 3. Write county.yaml - `adapter: generic`, `source.parcels` (path under raw/), optional `source.table` + keys. - `columns:` map standard fields to the county's columns. Leave out what doesn't exist. - `code_columns:` one column per code family (`state_class`, `land_use`, `building_type`...). - `codes:` map each relevant local code to a tag from `tag_glossary` in the trade's YAML. Only map codes you have a description for; unmapped codes are simply untagged. - `keyword_additions:` local public bodies (the state university, the port, the transit authority) under `PUBLIC`, local hospital systems under `HOSPITAL`, local tax consultants under `TAX_AGENT`. - `bbox: [west, south, east, north]` in lon/lat, from the parcel layer's bounds. - `nws_offices:` the NWS forecast office(s) covering the county (weather.gov shows it on the county's forecast page; e.g. HGX Houston, FWD Dallas–Fort Worth, EWX Austin/San Antonio). - `storm_events:` one or two defining storms for the region (look at the Iowa Environmental Mesonet LSR archive for the office and pick events with many damage reports) plus one `recent` event with `days: 365`. Methods are explained in `src/canopy_leads/storms.py`. - `downloads:` the direct file URLs so `canopy-leads fetch` can repeat the download next year. - `sources:` one row per source with its URL and what it's used for. ## 4. Test ```bash canopy-leads init runs/<county> --county path/to/county.yaml --trade tree_care canopy-leads fetch runs/<county> && canopy-leads normalize runs/<county> && canopy-leads classify runs/<county> ``` `normalize` prints the share of parcels with at least one tag; a very low share means codes are missing from the mapping. Review the checkpoint samples with the user before going on. If the county ships several linked tables that the generic adapter can't join, write a small adapter function in `src/canopy_leads/adapters.py` modelled on `hcad()`.
Wersje
W bazie nie ma innych skilli o nazwie add-county.
Migawka bazy z 13.11.2026 · wersja paczki 67396bf1a474 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100build-territory-list
Buduje ranking kont B2B dla firmy usługowej z publicznych danych: rejestrów podatkowych, zadrzewienia i raportów burz. - 100nws-flood-thresholds
Pobiera progi stanów powodziowych z NWS (National Weather Service): stany ostrzegawcze, niewielkie, umiarkowane i duże dla stacji USGS oraz dopasowuje stacje do progów. - 100tx-county-courts
Wskazuje właściwy sąd okręgowy lub sąd hrabstwa w Teksasie poza Harris i Dallas, z podziałem kompetencji między sądami.