What is this?
A collection of leaked system prompts from widely used LLM-based services, gathered as individual documents in one repository.
A collection of leaked system prompts from widely used LLM-based services, gathered as individual documents in one repository.Is this a tool or just a document collection?It is a plain document collection, not a tool or library; each entry is a text/markdown file containing a leaked prompt.
It is a plain document collection, not a tool or library; each entry is a text/markdown file containing a leaked prompt.How is the collection organized and how do I browse it?There is no install step; you just browse the repository files, where each leaked prompt is kept as a separate document following a consistent format.
There is no install step; you just browse the repository files, where each leaked prompt is kept as a separate document following a consistent format.Do I need anything to use this repo?No, it's just a set of text files you can read directly on GitHub; nothing to install or run.
No, it's just a set of text files you can read directly on GitHub; nothing to install or run.Why would a PR get rejected or the repo face takedown risk?The maintainer explicitly excludes sensitive commercial source code to avoid DMCA takedown, since the repo is cited in many academic papers.
The maintainer explicitly excludes sensitive commercial source code to avoid DMCA takedown, since the repo is cited in many academic papers.Why would someone want to read leaked system prompts?They reveal how real-world AI products are instructed to behave, which is useful for research, prompt engineering study, and understanding how popular LLM services are configured.
They reveal how real-world AI products are instructed to behave, which is useful for research, prompt engineering study, and understanding how popular LLM services are configured.Why is this repository cited in academic papers?Because it serves as a reference dataset of real-world system prompts, researchers use it to study prompt design and LLM behavior in production systems.
Because it serves as a reference dataset of real-world system prompts, researchers use it to study prompt design and LLM behavior in production systems.What are similar projects?awesome-chatgpt-prompts-zh, GPTs, chatgpt_system_prompt, system_prompts_leaks, prompts.chat, awesome-gpt-image-2-5-prompts, awesome-ai-research-writing, jailbreaks
awesome-chatgpt-prompts-zh, GPTs, chatgpt_system_prompt, system_prompts_leaks, prompts.chat, awesome-gpt-image-2-5-prompts, awesome-ai-research-writing, jailbreaksWhich topics is it in?system-prompt, prompt-library, prompts, document, prompt, collection, dataset, gpt, chatgpt, prompt-engineering, chatbot, awesome-list
system-prompt, prompt-library, prompts, document, prompt, collection, dataset, gpt, chatgpt, prompt-engineering, chatbot, awesome-list