A lot of AI systems give answers that look convincing, but reveal very little about how they were made. Footnote creates answers with a trail you can follow when it matters.
Transparency is important, but you also need control: over what it can and can't do, what's remembered, how much money can be spent, how carefully sensitive questions should be handled, and much more.
Footnote tries to bridge the gap between what AI can do and what people need in order to use it responsibly. See our philosophy page for the thinking behind the project.
Footnote is built with AI assistance. Read the AI use disclosure for more information.
Try Footnote in your browser with the live demo.
To run your own instance, follow the deployment guide. To contribute or run from the repository, follow Run from source.
Use this path to contribute to Footnote or run the repository directly.
Prerequisites:
- Node.js 22.13+
pnpm(pnpm@11.16.0+)
git clone https://github.com/footnote-ai/footnote.git
cd footnote
pnpm startpnpm start creates missing local setup files, installs dependencies when
needed, and starts the backend and web development services.
Footnote uses Docker containers for deployment on your own hardware or in the cloud. See the deployment guide for details.
Footnote is dual-licensed under MIT and HL3-CORE.