Skip to content

steepworksai/FormBuddy

Repository files navigation

FormBuddy — AI Form Filler

Fill any web form instantly from your personal documents — locally, privately, with AI-powered suggestions and full citations.

License: MIT Chrome Web Store Edge Add-ons Manifest V3


What it does

FormBuddy reads your PDFs, passport scans, tax forms, and notes — indexes them entirely on your device — then suggests the right value for every form field you click on any website.

  • Scan & Auto Fill — detects all fields on a page and fills them at once
  • Every suggestion cites its source — file name, page number, exact snippet
  • Works on any website — visa applications, insurance claims, tax returns, travel bookings
  • Multi-page form support — tracks which fields were already filled across page navigations
  • Quick Add — drag-drop files, right-click to save selected text, or take a screenshot

Privacy first

What happens Detail
Documents Read locally via File System Access API — never uploaded
LLM calls Short text snippets sent directly from your browser to your chosen AI provider
API keys Stored in Chrome's encrypted chrome.storage.local — never sent to FormBuddy
Analytics None. No backend server exists.

See the full Privacy Policy.


Bring Your Own Key (BYOK)

FormBuddy uses your own API key — no FormBuddy subscription or monthly fee.

Provider Models
Anthropic claude-sonnet-4-6, claude-haiku-4-5
OpenAI gpt-4o, gpt-4o-mini
Google gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.5-pro

Getting started

Install from a browser store

Add to Chrome → Add to Microsoft Edge →

Run locally from source

git clone https://github.com/steepworksai/FormBuddy.git
cd FormBuddy
npm install
npm run build

Then load the unpacked extension:

Chrome: Go to chrome://extensions → Enable Developer modeLoad unpacked → select dist/

Edge: Go to edge://extensions → Enable Developer modeLoad unpacked → select dist/


Tech stack

Layer Technology
Language TypeScript
Platform Chrome & Edge Extension — Manifest V3
Build Vite + @crxjs/vite-plugin
UI React + Tailwind CSS
PDF parsing pdfjs-dist + Tesseract.js (OCR)
LLM Anthropic SDK, OpenAI SDK, Gemini REST
Folder access File System Access API
Storage chrome.storage.local
Testing Vitest (188 unit tests) + Playwright (16 e2e tests)

Development

npm install          # install dependencies
npm run dev          # start Vite dev server
npm run build        # production build → dist/
npm run test:unit    # run Vitest unit tests
npm run test:e2e     # run Playwright e2e tests
npm run test:all     # build + all tests

Project structure

src/
├── background/      # Service worker — session state, LLM calls
├── content/         # Content script — field detection, autofill
├── sidepanel/       # Main UI — folder manager, scan & fill
├── popup/           # Settings — API key, model selection
└── lib/
    ├── llm/         # Claude, OpenAI, Gemini wrappers
    ├── indexing/    # Document indexing pipeline
    ├── parser/      # PDF text extraction + OCR
    └── folder/      # File System Access API helpers

Contributing

Issues and pull requests are welcome. Please open an issue first to discuss significant changes.

  1. Fork the repo
  2. Create a branch: git checkout -b feat/my-feature
  3. Make your changes with tests
  4. Open a pull request

License

MIT © 2026 SteepWorksAi

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors