Feature/add lightweight assistant

[emrgnt-cmplxty]

🚀	This description was created by Ellipsis for commit 729cb38

Summary:

Added lightweight assistant feature and refactored code for better modularity and maintainability.

Key points:

• Added Assistant, AssistantConfig, and Tool classes in r2r/base/abstractions/assistant.py.
• Updated r2r/base/__init__.py to include new assistant-related imports.
• Refactored r2r/main/api/routes to separate request models into individual files.
• Renamed several files for better organization, e.g., r2r/main/api/routes/auth.py to r2r/main/api/routes/auth/base.py.
• Updated r2r/providers/llm to rename LiteLLM and OpenAILLM to LiteLLMProvider and OpenAILLMProvider respectively.
• Modified r2r/main/app.py to use new route imports.
• Updated r2r/main/assembly/factory.py to use new LLM provider names.

---

Generated with ❤️ by ellipsis.dev

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to 729cb38 in 47.760326 seconds

More details

• Looked at 769 lines of code in 28 files
• Skipped 0 files when reviewing.
• Skipped posting 1 drafted comments based on config settings.

1. r2r/base/abstractions/assistant.py:55

• Draft comment:
  The implementation of the Assistant class and related components is well-structured and follows good OOP practices, providing a solid foundation for building conversational assistants.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The PR introduces a new module for an assistant abstraction, which includes classes like Assistant, AssistantConfig, Tool, Message, and Conversation. These classes are designed to facilitate the creation and management of conversational assistants that can interact with users and process responses using large language models (LLMs).

The Assistant class is an abstract base class that requires the implementation of run and process_llm_response methods, which are crucial for handling the assistant's main logic and processing responses from LLMs, respectively. The Conversation class manages message exchanges in a conversation, which is a typical requirement for chatbot-like functionalities.

The Tool class encapsulates a callable functionality that can be invoked by the assistant, potentially allowing the assistant to perform specific tasks or operations beyond simple message responses.

Overall, the PR seems to be well-structured and follows good OOP practices for extendability and maintenance.

Workflow ID: wflow_B01bNtB7btXtvBjK

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.