Open Election Data

Free data on who's running and where they stand. Every position is backed by an exact quote, checked against its source, that independent AI agents agreed on.

Endpoints

• /v1/index.json: what's here, counts and license
• /v1/issues.json: the 18 issues and how each side is worded
• /v1/all.json: every contest in one file
• /v1/contests/{id}.json: one contest
• /v1/divisions/{state}/{key}.json: contests for one district, for example /v1/divisions/ca/cd-10.json
• /schema.json: JSON Schema for a contest

Example

curl -s https://n8peace.github.io/open-election-data/v1/divisions/ca/cd-10.json

District keys follow U.S. Census districts: ca/state, ca/cd-10 (U.S. House), ca/sldu-9 (state senate), ca/sldl-15 (state house or assembly), ca/county-contra-costa, ca/place-oakland, ca/school-mount-diablo-unified.

What already exists

Good sources already exist: BallotReady offers candidate stances through a paid API, Vote Smart collects positions through its candidate survey, Ballotpedia tracks who's running, and Google Civic lists candidates near Election Day. This adds a free, open dataset of where candidates stand, with a checkable quote behind every position, free for anyone to reuse.

How it's made

Independent research agents on models from different AI companies each search and read on their own. A position is published only when they agree, and every quote is fetched from its source and matched word for word. Not finding a source never counts as a vote. Quotes are rechecked nightly. The method and all the code are on GitHub.

License

The data is under the Open Database License: use it for anything, including commercial products, as long as you credit Open Election Data and share improvements to the database under the same license. Quotes belong to the people and publications quoted.

Built for Plain Ballot, a free voter guide that uses this data.