An AI-powered database of bank account closure methods and their success rates, compiled from community-reported data on Doctor of Credit, updated daily. See an issue? Report it here.
Visit the site itself here.
Banks often offer sign-up bonuses for opening an account. Some people like to sign up for dozens of accounts and rake in the cash from all the sign-up bonuses. For these people, finding how to close an account quickly can be a legitimate time saver.
The leading source of how to close bank accounts is this Doctor Of Credit thread, but since it's manually updated, it often falls out of date.
scrape/
- Python scripts for scraping and processing bank closure data. This uses Gemini 2.5 Flash to convert unstructured comment data into structured data that the frontend consumesfrontend/
- SvelteKit web application - I vibe coded this. My intention was to learn Svelte, but the site was so simple that it was basically one-shottedcd frontend
npm install
npm run dev
Create a .env
file in the scrape/
directory:
GEMINI_API_KEY=your_api_key_here
cd scrape
uv pip install -e .
Scrape comments from Doctor of Credit:
cd scrape
python scrape.py
This creates/updates comments.jsonl
with the latest comments.
Extract structured data using AI:
python extract_data.py
This processes comments and generates comments_extracted.jsonl
and by_bank.json
.
Deploy the frontend:
cd frontend
npm run deploy
This builds and deploys the site to GitHub Pages.
MIT