Open data API

Datasets and endpoints

The base address is https://txbrb.muniopen.com/api/v1/. It lists every dataset, its row count, as-of date and download links. Read a dataset's metadata for its fields, definitions and the meaning of each blank.

Every response has public caching for 600 seconds, an ETag and Access-Control-Allow-Origin: *. Send the ETag in If-None-Match to receive HTTP 304 when unchanged. The API accepts GET only; other methods return HTTP 405. Browser access is allowed from any origin.

While a dataset is being rebuilt, its rows come from the last complete build, with that build's as_of, and are cached for 60 seconds. If no complete build exists, rows answer HTTP 503 with a Retry-After header.

Paging and filters

A rows response contains dataset, version, as_of, total, start, limit, next and rows. start is a zero-based offset. The default limit is 100; the maximum is 1000. Values above 1000 use 1000.

Follow next until it is null. It retains your year and issuer filters. Rows use a stable order for the same data version. A release can replace rows while you page; check the release calendar and restart your walk after a release if you need one complete snapshot.

Use ?fy=2025 for one fiscal year, or ?year=2026 for a program-year dataset. Texas fiscal years run from September 1 through August 31 and are named for the year they end. Metadata declares the supported year parameter. A dataset with an issuer_field also accepts issuer, the published stable issuer id, rather than an issuer name. Other filters are unsupported.

Unknown or repeated parameters, non-integer page or year values, negative offsets, empty issuer ids and limits below 1 return HTTP 400 with a JSON error. Unknown datasets return HTTP 404. The grids' ?format=json responses serve the user interface and have a separate contract.

Field names are the stable API keys in the dictionary. Money is a decimal string, such as "1250000.00", with no thousands separators; parse it with a decimal type to preserve cents. Dates use YYYY-MM-DD. Missing values are JSON null, never zero; each field's dictionary explains what a blank means.

Copy-paste examples

curl

Get the first page of local issues for fiscal year 2025:

curl --fail 'https://txbrb.muniopen.com/api/v1/datasets/local-issues/rows/?fy=2025&start=0&limit=1000'

Python

This uses only Python's standard library and follows every page. Use Decimal when reading money strings.

import json
from decimal import Decimal
from urllib.request import urlopen
from urllib.parse import urljoin

base = "https://txbrb.muniopen.com/api/v1/"
url = base + "datasets/local-issues/rows/?fy=2025&limit=1000"
rows = []
while url:
    with urlopen(url) as response:
        page = json.load(response)
    rows.extend(page["rows"])
    url = urljoin(base, page["next"]) if page["next"] else None
principal = sum((Decimal(row["principal"]) for row in rows
                 if row["principal"] is not None), Decimal("0"))
print(len(rows), principal)

Excel Power Query

In Excel, choose Data → Get Data → From Other Sources → Blank Query, open Advanced Editor and paste this query. It follows every page and leaves money as exact text; choose your decimal conversion deliberately.

let
    Base = "https://txbrb.muniopen.com/api/v1/",
    GetPage = (url as text) as record => Json.Document(Web.Contents(url)),
    First = Base & "datasets/local-issues/rows/?fy=2025&limit=1000",
    Pages = List.Generate(
        () => GetPage(First),
        each _ <> null,
        each if [next] = null then null
             else GetPage(Uri.Combine(Base, [next])),
        each [rows]
    ),
    Rows = List.Combine(Pages),
    Data = Table.FromRecords(Rows)
in
    Data

Compatibility policy

Inside v1, fields are only added, never renamed or removed. A breaking change becomes v2, and v1 keeps running for at least 12 months after v2 is released. Every field addition is listed below with its date. Applications should accept added fields and check a field's blank meaning before treating missing data.

Change log

10/6/2026 — Version 1 published

The initial version added the dataset index, metadata and paged rows. The field additions are listed individually: