About OmnibusAI

What is this?

OmnibusAI is a free, automated service that generates plain-English breakdowns of federal omnibus appropriations bills. When Congress passes a massive spending bill — often 1,000+ pages of dense legislative text — we use AI to read it section by section and explain what it means in everyday language.

How does it work?

  1. Monitoring: We automatically check Congress.gov for new appropriations and omnibus bills.
  2. Fetching: When a bill is found, we download its full text in XML format from GovInfo.gov (the Government Publishing Office's public repository).
  3. Parsing: The XML is parsed into its natural hierarchy — divisions (departments), titles (program areas), and sections (individual provisions). Dollar amounts are extracted from the text.
  4. Summarizing: Each section is sent to an AI language model with a prompt asking for a plain-English explanation, key spending figures, and notable provisions.
  5. Publishing: Summaries are compiled into this website, organized by department and program area.

Why does this exist?

Omnibus spending bills are some of the most impactful legislation Congress passes — they fund the entire federal government. But they're also some of the most unreadable: thousands of pages of legal text, cross-references, and bureaucratic language.

Existing tools for understanding these bills are either paid (aimed at lobbyists and policy professionals) or sparse (one-off summaries by news outlets). We believe everyone deserves access to clear, comprehensive breakdowns of how their tax dollars are being spent.

Important disclaimers

  • AI-generated content: All summaries are generated by AI language models. While we strive for accuracy, AI can make mistakes, misinterpret provisions, or miss nuances. Always verify important information against official sources.
  • Not legal advice: This site provides educational summaries, not legal or financial advice.
  • Non-partisan: We present information factually without political commentary. Our AI prompts are designed to be balanced and non-editorial.
  • Public domain source: All bill text is public domain, published by the U.S. Government. Our summaries are original content.

Data sources

Open source

This project is open source. You can view the code, report issues, or contribute on GitHub.