🔬 Multi-Agent Research Assistant

A fully local, privacy-first research assistant powered by LangGraph (multi-agent orchestration), Ollama (local LLMs), and Streamlit (frontend UI).

---

Architecture

User Query
    │
    ▼
┌─────────────┐     ┌──────────────┐     ┌─────────────┐     ┌────────────┐
│   PLANNER   │────▶│  RESEARCHER  │────▶│   ANALYST   │────▶│   WRITER   │
│             │     │              │     │             │     │            │
│ Breaks query│     │ Calls tools  │     │ Synthesizes │     │ Produces   │
│ into a plan │     │ to gather    │     │ findings &  │     │ polished   │
│             │     │ information  │     │ insights    │     │ report     │
└─────────────┘     └──────────────┘     └─────────────┘     └────────────┘
                           │
                    ┌──────┴───────┐
                    │    TOOLS     │
                    ├──────────────┤
                    │ web_search   │  DuckDuckGo / Tavily
                    │ wikipedia    │  Wikipedia API
                    │ arxiv        │  arXiv papers
                    │ scrape_page  │  BeautifulSoup
                    │ calculator   │  Safe math eval
                    │ summarize    │  Ollama summarizer
                    └──────────────┘

---

Quick Start

1. Install Ollama

# macOS / Linux
curl -fsSL https://ollama.com/install.sh | sh

# Pull a model 
ollama pull llama3.2:1b

2. Install Python dependencies

cd research_assistant_agent
python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate

pip install -r requirements.txt

3. Configure (optional)

cp .env.example .env
# Edit .env to set your model, Tavily API key, etc.

4. Run the app

streamlit run app.py

Open http://localhost:8501 in your browser.

---

Tools

Tool	Description	Requires
web_search	Searches the web (Tavily → DuckDuckGo fallback)	duckduckgo-search
wikipedia_search	Looks up Wikipedia articles	wikipedia
arxiv_search	Searches arXiv for academic papers	arxiv
scrape_webpage	Fetches and extracts text from URLs	beautifulsoup4
calculator	Evaluates math expressions safely	built-in
summarize_text	Summarizes text via Ollama	Ollama

For higher-quality web search, get a free API key at tavily.com and add it in the sidebar.

---

Project Structure

research_assistant/
├── app.py                  # Streamlit frontend
├── graph.py                # LangGraph pipeline definition
├── requirements.txt
├── .env.example
├── config/
│   └── settings.py         # Central config (reads .env)
├── agents/
│   ├── state.py            # Shared ResearchState TypedDict
│   ├── planner.py          # Planner agent
│   ├── researcher.py       # Researcher agent (tool-calling loop)
│   ├── analyst.py          # Analyst agent
│   └── writer.py           # Writer agent
└── tools/
    └── research_tools.py   # All tool definitions

---