trun your raycast pro to a OpenAI API compatible api server

you can use docker or run it directly

if you have Docker installed, you can run this command to start a server

docker run -dit --name raychat \
	-p 8080:8080 \
	-e EMAIL='your_email' \
	-e PASSWORD='your_password' \
	-e CLIENT_ID='your_client_id' \
	-e CLIENT_SECRET='your_client_secret' \
	-e EXTERNAL_TOKEN='your_fake_openai_token' \ # you can provide multi token like "token_a,token_b", token splitted with comma
	--restart always \
	vaalacat/raychat:latest

or if you already have a token, you can run this command

docker run -dit --name raychat \
	-p 8080:8080 \
	-e TOKEN='your_token' \
	-e EXTERNAL_TOKEN='your_fake_openai_token' \ # you can provide multi token like "token_a,token_b", token splitted with comma
	--restart always \
	vaalacat/raychat:latest

then you can use http://localhost:8080/v1/chat/completions to test your server, arm and amd64 are both supported

0. clone this repo and cd into it

1. Install Raycast

2. Set Fiddler to capture traffic, get the ClientID and ClientSecret

3. Put the ClientID and ClientSecret into a .env file, fill Email and Password

4. Run go run main.go your server will start at http://localhost:8080

you can use http://localhost:8080/v1/chat/completions to test your server