sshh12/llm_osint — reverse-engineered prompt
Reverse engineered prompt
Build me a small Python proof of concept that uses an LLM to research someone on the public internet and then do a task with what it finds. I want it to feel simple, like I can run one command with a person’s name and a question, and it will search the web, pull together the most useful public details, and answer in a clean readable format.
The main demo should be a person lookup flow. For example, I should be able to give it a name like “Shrivu Shankar” and ask it to create a markdown resume, write a short profile, or summarize likely strengths and weaknesses based only on public sources. If the name is common, let me add a little extra context to disambiguate.
Please keep it clearly labeled as a proof of concept, include a few example scripts, and add a note about privacy and ethical use since this kind of tool can get invasive fast. Make setup and usage easy, and look up current docs online if you need to.
Want more depth? Deep Reverse