House Clerk roll-call votes

planned Legislative · Tier 1 · XML index · Clerk of the House of Representatives

Official site (opens in a new tab) · All sources

What this source is

Every House roll-call vote as XML at clerk.house.gov/evs/{year}/roll{NNN}.xml with a yearly index — published same-day, examples back to 2001. Pairs with the Congress.gov API's beta house-vote endpoint as a cross-check.

Identity and registry record

Registry id

house-clerk-votes

Agency / parent organization

Clerk of the House of Representatives

Branch

legislative

Type

XML index

Status

planned

Tier

1

URL (home)

https://clerk.house.gov/Votes (opens in a new tab)

URL (index)

https://clerk.house.gov/evs/2026/index.asp (opens in a new tab)

Registered

2026-07-28

Registry notes

Vote XML files verifiably exist and are indexed (2026-07-28). The formal schema documentation lives on xml.house.gov, which 403'd research fetchers — read it via our identified client at probe. 2026-07-31, live: the per-vote XML is confirmed (evs/2026/roll283.xml → 200 text/xml, 82,500 bytes, full metadata + member positions) and clerk.house.gov publishes NO robots.txt (404, nothing disallowed), but the YEAR INDEX IS HTML, NOT XML — evs/2026/index.asp is a 7 KB <TABLE> of ~15 recent votes linking to cgi-bin/vote.asp, evs/2026/index.xml is 404, and clerk.house.gov/Votes is a 249 KB JavaScript application. So this entry needs the html-index adapter (plan Phase 5), not the xml-index one that activated senate-xml; it stays planned rather than ship half-understood. Type stays xml-index because the per-vote records are XML; revisit the type when the index adapter lands.

How we ingest it

Channel

XML index

Method

Would walk the numeric roll-call sequence from a watermark, fetching new vote XML files.

Request budget

the agency class: at most 1,500 requests per day shared across every agency web source, counted from the fetch log (failed requests count too)

Politeness

robots.txt is honored as observed — including each host's crawl-delay, exactly — and every request identifies itself as fapd/0.1 (Free Agentic Publication Digester; +https://fapd.info/bot.html; contact: hustleyourcity@gmail.com); a refusal is recorded, never evaded

Capture and hash

captured raw content is hashed (SHA-256) into the day's committed provenance manifest, hash-chained day to day (PROVENANCE.md)

Ingestion health

Not ingested: the registry status of this source is planned.

Ingestion statistics

These figures describe this project's ingestion of this source — items we recorded and requests we made — and nothing else. They are not a measurement of the publisher.

Not ingested: the registry status of this source is planned. Ingestion statistics are measured for active sources only.

All time

Last 30 days, day by day

No requests and no items were recorded in the last 30 days, so there is nothing to chart.