Merge pull request #50 from ghost-in-moss/dev

v0.1.0 new beta preparing

README.md

@@ -65,8 +65,16 @@ ghostos init
 ```
 
 Configure the model. Default to use OpenAI `gpt-4o`, requiring the environment variable `OPENAI_API_KEY`.
-Or you can use configuration ui by streamlit:
 
+```bash
+export OPENAI_API_KEY="your openai api key"
+# Optionals: 
+export OPENAI_PROXY="sock5://localhost:[your-port]" # setup openai proxy
+export DEEPSEEK_API_KEY="your deepseek api key"
+epoxrt MOONSHOT_API_KEY="your moonshot api key"
+```
+
+Or you can use configuration ui by streamlit:
 ```bash
 ghostos config
 ```
@@ -85,11 +93,32 @@ that can be instructed to call functions or methods within the file through natu
 ghostos web [my_path_file_path]
 ```
 
-ou can create a local Python file and define your own Agents. For more details
+some demo agents
+```bash
+ghostos web ghostos.demo.agents.jojo
+ghostos web ghostos.demo.test_agents.moonshot         # moonshot-v1-32k model
+ghostos web ghostos.demo.test_agents.deepseek_chat    # deepseek chat model
+ghostos web ghostos.demo.test_agents.openai_o1_mini   # openai o1 mini model
+```
+
+You can create a local Python file and define your own Agents. For more details
 
 * [Chatbot](docs/zh-cn/usages/chatbot.md): simplest chatbot
 * [MossAgent](docs/zh-cn/usages/moss_agent.md): an agent that can interact with the python module
 
+## Install Realtime 
+
+`GhostOS` support [OpenAI Realtime](https://platform.openai.com/docs/guides/realtime),
+using [pyaudio](https://pypi.org/project/PyAudio/) to handle realtime audio i/o. 
+Need to install the dependencies first:
+
+```bash
+pip install 'ghostos[realtime]'
+```
+
+> You may face some difficulties while install pyaudio on your device,
+> I'm sure gpt-4o, google or stackoverflow will offer you solutions.
+
 ## Use In Python
 
 ```python

docs/en/README.md

@@ -66,6 +66,15 @@ ghostos init
 ```
 
 Configure the model. Default to use OpenAI `gpt-4o`, requiring the environment variable `OPENAI_API_KEY`.
+
+```bash
+export OPENAI_API_KEY="your openai api key"
+# Optionals: 
+export OPENAI_PROXY="sock5://localhost:[your-port]" # setup openai proxy
+export DEEPSEEK_API_KEY="your deepseek api key"
+epoxrt MOONSHOT_API_KEY="your moonshot api key"
+```
+
 Or you can use configuration ui by streamlit:
 
 ```bash
@@ -86,11 +95,33 @@ that can be instructed to call functions or methods within the file through natu
 ghostos web [my_path_file_path]
 ```
 
-you can create a local Python file and define your own Agents. For more details
+some demo agents
+
+```bash
+ghostos web ghostos.demo.agents.jojo
+ghostos web ghostos.demo.test_agents.moonshot         # moonshot-v1-32k model
+ghostos web ghostos.demo.test_agents.deepseek_chat    # deepseek chat model
+ghostos web ghostos.demo.test_agents.openai_o1_mini   # openai o1 mini model
+```
+
+You can create a local Python file and define your own Agents. For more details
 
 * [Chatbot](/en/usages/chatbot.md): simplest chatbot
 * [MossAgent](/en/usages/moss_agent.md): an agent that can interact with the python module
 
+## Install Realtime
+
+`GhostOS` support [OpenAI Realtime](https://platform.openai.com/docs/guides/realtime),
+using [pyaudio](https://pypi.org/project/PyAudio/) to handle realtime audio i/o.
+Need to install the dependencies first:
+
+```bash
+pip install 'ghostos[realtime]'
+```
+
+> You may face some difficulties while install pyaudio on your device,
+> I'm sure gpt-4o, google or stackoverflow will offer you solutions.
+
 ## Use In Python
 
 ```python

docs/en/getting_started/installation.md

@@ -37,6 +37,14 @@ natural language dialogue.
 ghostos web [my_path_file_path]
 ```
 
+## Install Realtime
+
+install realtime dependencies (mainly `pyaudio`):
+
+```bash
+pip install 'ghostos[realtime]'
+```
+
 ## Workspace
 
 `GhostOS` is currently using local files to store runtime data, so it's necessary to initialize a workspace.
@@ -56,7 +64,8 @@ ghostos clear-runtime
 
 There are two ways to define these environment variables:
 
-Using a `.env` file (automatically read through `dotenv`)
+* Export the environment variables to your shell.
+* Using a `.env` file (automatically read through `dotenv`)
 
 ```bash
 

docs/zh-cn/README.md

@@ -59,7 +59,16 @@ ghostos init
 ```
 
 配置大模型. 默认使用 OpenAI `gpt-4o`, 要求环境变量存在 `OPENAI_API_KEY`.
-或者运行 `streamlit` 编辑界面:
+
+```bash
+export OPENAI_API_KEY="your openai api key"
+# Optionals: 
+export OPENAI_PROXY="sock5://localhost:[your-port]" # setup openai proxy
+export DEEPSEEK_API_KEY="your deepseek api key"
+epoxrt MOONSHOT_API_KEY="your moonshot api key"
+```
+
+或者运行 `streamlit` 打开配置界面:
 
 ```bash
 ghostos config
@@ -78,11 +87,32 @@ ghostos web ghostos.demo.agents.jojo
 ghostos web [my_path_file_path]
 ```
 
+当前的测试用例:
+
+```bash
+ghostos web ghostos.demo.agents.jojo
+ghostos web ghostos.demo.test_agents.moonshot         # moonshot-v1-32k model
+ghostos web ghostos.demo.test_agents.deepseek_chat    # deepseek chat model
+ghostos web ghostos.demo.test_agents.openai_o1_mini   # openai o1 mini model
+```
+
 可以通过创建本地 python 文件, 定义出自己的 Agents. 详情请见:
 
 * [Chatbot](/zh-cn/usages/chatbot.md): 极简的对话机器人
 * [MossAgent](/zh-cn/usages/moss_agent.md): 能使用 python 的 agent
 
+## Install Realtime
+
+`GhostOS` 当前支持 [OpenAI Realtime](https://platform.openai.com/docs/guides/realtime),
+使用 [pyaudio](https://pypi.org/project/PyAudio/) 来处理实时语音的输入输出.
+需要安装相关依赖: 
+
+```bash
+pip install 'ghostos[realtime]'
+```
+
+> 在安装 pyaudio 的时候可能会遇到一些问题, 我相信 gpt-4o, google 和 stackoverflow 能够很好地帮助你解决它们. 
+
 ## Use In Python
 
 ```python

docs/zh-cn/getting_started/installation.md

@@ -34,6 +34,14 @@ ghostos web ghostos.demo.agents.jojo
 ghostos web [my_path_file_path]
 ```
 
+## Install Realtime
+
+安装 realtime 所需的依赖: 
+
+```bash
+pip install 'ghostos[realtime]'
+```
+
 ## Workspace
 
 `GhostOS` 当前版本使用本地文件来存运行时数据. 所以需要初始化一个 workspace.
@@ -51,6 +59,7 @@ ghostos clear-runtime
 `GhostOS` 依赖各种模型的 `access token`, 默认是从环境变量中读取.
 定义这些环境变量有两种方法:
 
+- export 环境变量到命令行中. 
 - 使用 `.env` 文件 (自动通过 `dotenv` 读取)
 
 ```bash

ghostos/__main__.py

@@ -0,0 +1,4 @@
+from ghostos.scripts.cli import main
+
+if __name__ == "__main__":
+    main(prog_name="help")