izaart95-jpg/KimiFreeAPI ? reverse-engineered prompt
Reverse engineered prompt
Build me a small Go service that acts like an OpenAI compatible proxy for Kimi AI.
I want to run it locally with one access token, then send chat completion requests to it from any app that already talks to OpenAI. It should support normal chat responses and streaming over SSE, handle both simple text and richer message content, and let me turn conversation history on or off. It should also fetch the available Kimi models automatically, let me switch the default model, and refresh the model list without restarting.
Please add a simple auth check for incoming requests, a way to start a fresh chat session, and clean shutdown when the server stops. It should be fast and lightweight, with good connection reuse and no extra dependencies if possible. If you need current details for Kimi’s API shape, look them up online and match the real behavior closely.
Are you gonna build this?
make sure you review the code using coderabbit