use litellm underlying

add some logic in workspace

introduce coding assistant ui

rename project to client-app

reactor ui project

add chinese voice if detecting chinese

add a simple ui for testing voice-voice chat

add gcloud_tts

introduce gcloud tts

add demo code & cors enabled & document the apis

fix typo

include stt integration test

introduce stt endpoint

refactor the routes api to be more modularized

add stt support

poc proven for stt

add support of streaming TTS

add tts support from user end

refactor the integration test but tts is broken

expose ttls model for external configuration

introduce gemini tts model into the system

add support for multi-models

refactor retriever

refactor the services to be plugable

add db path support in yaml config

fine-tune the vector store for extension

add faiss support in bash script

import real gemini embeddings

add toggle on client side to enable or disable faiss db retriever

add log level in the config

env prioritize than yaml input

support switch model in between

add readme file

support chat history

introduce the session into the frontend api

introduce the delete document funcs

refactor the config input

introduce the yaml for configuration

further refactor

fix typo

refactor db creation & centralize the config.py

refactor

refactor code

introduce chain of throught

fix some issues

add dspy

Refactor core structure and add new RAG modules

Add database and model to the project

Refactor the code structure to allow using both gemini and Deepseek.

mark milestone first done

first commit