routerhubSI/deepseek-uncensored ? reverse-engineered prompt
Reverse engineered prompt
Build me a clean documentation site and reference repo for DeepSeek uncensored and abliterated builds, mainly for security research and red teaming.
I want a homepage that explains the different model variants in plain language, shows which one to use for pinned reproducible runs versus the rolling build, and includes simple pricing and context window info. Add individual model pages, a quickstart for using the hosted API, and examples for running the GGUF builds locally with Ollama or llama.cpp. Please also include a red team prompt library page, a guide for testing models against refusal behavior, and a few copy paste examples for OpenAI compatible and Anthropic compatible usage.
Make it feel polished and easy to skim, with clear navigation, good formatting, and callouts for the main recommended build. If anything depends on current provider docs, look them up online so the setup steps are accurate.
Are you gonna build this?
make sure you review the code using coderabbit