yeet ur data into the SWAMP.
clean UI comes out. no cap.

webhook, JSON, CSV, that cursed spreadsheet — throw it in. SWAMP reads it, types every column, and speedruns you a grid, board, gallery and dashboard. it just works.

POST /api/datasets?name=Orders
$ curl -X POST \
   '…/api/datasets?name=Orders' \
   -H 'content-type: application/json' \
   -d '{ "data": [
        { "order_id": "ORD-7001",
          "total": 38.49,
          "status": "paid" }
   ] }'
200cookedGridBoardGalleryDashboard

what it do

it just knows

emails, money, dates, statuses, links, pics — 13 column types sniffed straight from the values. zero schema mapping. big brain energy.

ur edits are sacred

rename or hide fields and it sticks. re-imports can never overwrite your changes. respectfully.

views spawn themselves

grid, board, gallery, dashboard — they appear when the data supports them. edit statuses inline, drag cards between lanes. goes brrr.

eats everything

CSV, XLSX, JSON — or POST raw webhook chaos at the API and the dataset shows up ready to use. om nom.

the lore

1

yeet data

upload a file or POST JSON to /api/datasets. nested webhook payloads flatten themselves. fr.

2

engine cooks

every column gets a semantic type, a field registry entry, and a confidence score. no thoughts, just types.

3

you vibe

sort, search, filter the grid. drag kanban cards. read the dashboard that built itself. gg.

3 sample datasets already in there. free real estate.

open the workspace, poke a status, drag a card, filter the grid. touch data, not grass.

enter the swamp