Merge pull request #52 from filip-michalsky/update_docs_and_tests

Update docs and tests

CHANGELOG.txt

@@ -1,5 +1,11 @@
 Updates to the SalesGPT project: Building the world's best AI Sales Agents.
 
+September 8 2023
+---------------
+Version 0.0.7
+- SalesGPT is now compatible with LiteLLM - choose any closed/open-sourced LLM 
+to work with SalesGPT.
+
 August 23 2023
 ---------------
 Version 0.0.6

README.md

@@ -20,6 +20,7 @@ Please send an email to [the repo author](mailto:[email protected]).
 ## :red_circle: Latest News
 
 - Sales Agent can now take advantage of **tools**, such as look up products in a product catalog!
+- SalesGPT is now compatible with [LiteLLM](https://github.com/BerriAI/litellm), choose any closed/open-sourced LLM to work with SalesGPT! Thanks to LiteLLM maintainers for this contribution!
 
 ### Demo: Outbound Prospecting from Crusty AI: A New Way to Sell? 🤔
 
@@ -37,7 +38,7 @@ Please send an email to [the repo author](mailto:[email protected]).
 ```python
 import os
 from salesgpt.agents import SalesGPT
-from langchain.chat_models import ChatOpenAI, ChatLiteLLM
+from langchain.chat_models import ChatLiteLLM
 
 os.environ['OPENAI_API_KEY'] = 'sk-xxx' # fill me in
 
@@ -128,7 +129,7 @@ As such, this agent can have a natural sales conversation with a prospect and be
 
 ## Installation
 
-Make sure your have a python 3.10+ and run:
+Make sure your have a **python 3.10+** and run:
 
 `pip install -r requirements.txt`
 
@@ -165,14 +166,15 @@ Follow me at [@FilipMichalsky](https://twitter.com/FilipMichalsky)
 ## SalesGPT Roadmap
 
 - [high priority] Sell your soul.
-- [high priority] Add support for multiple LLMs backends [PR in progress here](https://github.com/filip-michalsky/SalesGPT/pull/36)
 - [high priority] Improve reliability of the parser [issue here](https://github.com/filip-michalsky/SalesGPT/issues/26) and [here](https://github.com/filip-michalsky/SalesGPT/issues/25)
 - Add example implementation of OpenAI functions agent[issue here](https://github.com/filip-michalsky/SalesGPT/issues/17)
 - Add support for multiple tools [issue here](https://github.com/filip-michalsky/SalesGPT/issues/10)
-- Add an agent controller for whne stages need to be traversed linearly without skips [issue here](https://github.com/filip-michalsky/SalesGPT/issues/19)
-- Add `tool_getter` to choose a tool based on vector distance to the taks needed to be done
+- Add an agent controller for when stages need to be traversed linearly without skips [issue here](https://github.com/filip-michalsky/SalesGPT/issues/19)
+- Add `tool_getter` to choose a tool based on vector distance to the tasks needed to be done
 - What tools should the agent have? (e.g., the ability to search the internet)
 - Add the ability of Sales Agent to interact with AI plugins on your website (.well-known/ai-plugin.json)
+
+~~-- [high priority] Add support for multiple LLMs backends [PR in progress here](https://github.com/filip-michalsky/SalesGPT/pull/36)~~-
 ~~-
  Add the ability to stop generation when user interupts the agent~~
 

examples/streaming_generator_example.py

@@ -4,13 +4,8 @@
 
 from salesgpt.agents import SalesGPT
 
-with open(".env", "r") as f:
-    env_file = f.readlines()
-envs_dict = {
-    key.strip("'"): value.strip("\n")
-    for key, value in [(i.split("=")) for i in env_file]
-}
-os.environ["OPENAI_API_KEY"] = envs_dict["OPENAI_API_KEY"]
+from dotenv import load_dotenv
+load_dotenv()
 
 llm = ChatLiteLLM(temperature=0.9)
 

salesgpt/salesgptapi.py

@@ -1,6 +1,6 @@
 import json
 
-from langchain.chat_models import ChatOpenAI, ChatLiteLLM
+from langchain.chat_models import ChatLiteLLM
 
 from salesgpt.agents import SalesGPT
 

salesgpt/version.py

@@ -1,3 +1,3 @@
 """Version information."""
 
-__version__ = "0.0.6"
+__version__ = "0.0.7"

tests/test_salesgpt.py

@@ -119,7 +119,8 @@ def test_valid_inference_stream(self, load_env):
 
     @pytest.mark.asyncio
     async def test_valid_async_inference_stream(self, load_env):
-        llm = ChatOpenAI(temperature=0.9)
+
+        llm = ChatLiteLLM(temperature=0.9)
         model_name = "gpt-3.5-turbo"
 
         sales_agent = SalesGPT.from_llm(