Automated AI Research Assistant

This project is a Streamlit application that uses CrewAI to automate the process of researching a given topic, generating a structured summary based on arXiv research papers, and allowing users to ask follow-up questions on the generated summary.

It leverages the power of Large Language Models (LLMs) via the langchain-google-genai library (specifically with Gemini) and custom tools to create a collaborative team of AI agents for research and summarization.

---

Features

• Automated Research: A multi-agent crew (Research Assistant and Research Summarizing Writer) works together to research a topic.
• arXiv Integration: A custom arxiv_search tool fetches relevant, up-to-date research papers.
• Structured Summaries: Generates detailed, multi-paragraph research summaries, complete with a references section.
• Interactive Q&A: Allows users to ask follow-up questions directly on the generated summary.
• Downloadable Output: The research summary is available for download as a markdown file.
• Streamlit Interface: A clean and interactive web interface for a smooth user experience.

---

Prerequisites

Before running the application, ensure you have the following installed:

1. Python 3.8+
2. A Gemini API Key.

---

Getting Started

1. Setup Environment

Clone the repository (assuming you have your project files in a single directory):

# Clone your project repository (if applicable)
# git clone <your-repo-link>
# cd automated-ai-research-assistant

2. Install Dependencies

Install the required Python packages using the requirements.txt file:

pip install -r requirements.txt

3. Configure API Key

The project requires a Gemini API key for the LLM.

1. Open the .env file.

2. Replace the placeholder with your actual key:

    GEMINI_API_KEY="YOUR_GEMINI_API_KEY_HERE"
    CREWAI_TRACING_ENABLED=true

> **Note:** The `llm` is configured in `agents.py` to use `gemini-2.5-flash-lite`.

4. Run the Application

Start the Streamlit application from your terminal:

streamlit run app.py

This will open the application in your web browser.

---

How to Use

1. Enter Topic: In the sidebar, enter a research topic (e.g., "Augmented Reality in Healthcare").
2. Run Research: Click the "Run Research" button. The CrewAI process will start, and a loading spinner will appear.
3. View Summary: Once complete, the detailed research summary will appear in the main column.
4. Ask Questions: Use the "Ask Questions" section to query the summary with follow-up questions.
5. Download: Click "Download Summary" to save the output locally.
6. Clear: Click "Clear Summary" to start a new research session.

---

Project Structure

File	Description
app.py	The main Streamlit application, handling the UI and crew.kickoff.
crew.py	Defines the CrewAI Crew with the agents and tasks.
agents.py	Defines the LLM and the two CrewAI Agents (topic_researcher, topic_writer).
tasks.py	Defines the CrewAI Tasks for research and writing, including the expected output and file saving.
tools.py	Contains the custom ArxivSearchTool for fetching papers from arXiv.
requirements.txt	Lists all necessary Python dependencies.
style.css	Custom CSS for the Streamlit application's styling.
.env	Environment file for storing the GEMINI_API_KEY.

---

CrewAI Agents and Tasks

Agents

• Research Assistant (topic_researcher):

  • Role: Expert in finding relevant information from research papers.
  • Goal: Research the given topic from research papers, focusing on process, methodology, reasons, and results.
  • Tools: arxiv_search_tool

• Research Summarizing Writer (topic_writer):

  • Role: Expert in writing detailed, well-structured research summaries.
  • Goal: Write a detailed research summary about the topic, including references.
  • Tools: arxiv_search_tool (Used for referencing/validation)

Tasks

1. Research Task: Delegates the search to the Research Assistant.
2. Write Task: Delegates the writing to the Research Summarizing Writer.
   • Expected Output: A 5-paragraph detailed research summary formatted as markdown, including a references section.
   • Output File: research_summary1.md

The crew is configured to run in Process.sequential mode, meaning the Research Assistant performs its task first, and then the Research Summarizing Writer uses those findings to complete the writing task.