How mrwadams/takedown-gpt works
Project type and purpose
• This is a single-page Streamlit web application (app.py) whose purpose is to help users draft takedown-request emails to domain registrars.
• It uses LangChain agents orchestrating OpenAI’s function-calling capability and interacts with external services (WHOIS/RDAP lookup, DuckDuckGo search).
File structure
• Top-level files only:
• app.py (the entire UI + business logic)
• requirements.txt (dependencies: Streamlit, OpenAI SDK v1, LangChain, python-whois or rdap client, duckduckgo-search, etc.)
• README.md, .gitignore
• No separate frontend/backend folders: everything runs in one Python process (monolith).
README for technical insights
• Core flow:
a) User enters domain, OpenAI API key, selects GPT model & lookup protocol
b) App uses WHOIS/RDAP to identify the registrar
c) App uses DuckDuckGo to find takedown-request email address
d) LangChain agent + function calling draft the email
e) Streamlit displays the draft for copy/download
• Key technologies: Streamlit, OpenAI Python SDK, LangChain, WHOIS/RDAP, DuckDuckGo, Python.