AI Chatbot & Autonomous Researcher

A full-stack prototype that pairs a conversational AI with an autonomous background-research engine. It ships with:

• React 19 front-end
• FastAPI + LangGraph back-end
• Motivation-driven research scheduler
• Privacy-preserving AI personalization system
• Knowledge Graph visualization (powered by Zep)
• JWT-secured admin console & prompt editor

Quick Start (dev)

One-shot (recommended)

# install JS dev deps (only needed once)
npm install

# start both backend (Uvicorn ‑-reload) + frontend (React) in parallel
npm run dev

That command:

1. activates the virtual-env inside backend/venv/.
2. launches Uvicorn with auto-reload and a 20-second graceful-shutdown window.
3. starts the React dev server.
4. streams both logs side-by-side via concurrently.

Manual (legacy)

# 1. clone repo & run helper script
./setup.sh               # installs backend venv + npm deps

# 2. start backend
cd backend && source venv/bin/activate
uvicorn app:app --reload

# 3. start frontend (new shell)
cd ../frontend && npm start

Detailed installation / production guides live in the docs folder.

Repository Layout

backend/   FastAPI application, LangGraph, storage
frontend/  React UI
setup.sh   one-shot installer for dev stacks
docs/      extended documentation (setup, research engine, admin debug …)

Key Features

🤖 AI Personalization System

The assistant learns your preferences and adapts its responses while keeping all data private and transparent:

• Privacy-First: All personalization data stored locally, never transmitted externally
• Adaptive Learning: Learns from your reading patterns, source preferences, and interaction habits
• Transparent Control: See exactly what the system has learned with full override capabilities
• Smart Formatting: Automatically adjusts response length, detail level, and structure
• Source Intelligence: Prioritizes research sources based on your engagement patterns

👤 User Experience: Three-tab interface (Personality, Preferences, What I've Learned) gives you complete control over how the AI adapts to your needs.

🔬 Multi-Source Research Engine

Motivation-driven background research that continuously learns about topics you're interested in. Features intelligent multi-source search that combines web, academic, social (Hacker News), and medical sources for comprehensive responses.

Key Features:

• Autonomous Topic Expansion - System automatically discovers related topics using knowledge graph analysis and AI
• Smart Limits - Enforces maximum active topics (default: 10) to prevent overwhelming users
• Graceful Handling - When at capacity, expansion topics are created as inactive suggestions for manual review
• Manual Control - Create custom research topics or choose from AI-suggested expansions

📊 Knowledge Graph Visualization

Visual representation of research findings and topic relationships powered by Zep memory.

Learn More

• docs/setup.md – prerequisites, env variables, testing
• docs/research_engine.md – how autonomous research & motivation work
• docs/personalization.md – AI personalization system (complete guide + API reference)
• docs/ARCHITECTURE.md – multi-source search system architecture
• docs/admin_debugging.md – admin UI, debug APIs, logging & tracing
• docs/user_guide.md – everyday usage including Knowledge Graph
• docs/zep_graph_visualization.md – Knowledge Graph implementation details
• docs/troubleshooting.md – common errors and fixes

License

GPL-3.0