🚀 AgentifyX

Transform conventional software systems into Agentic AI architectures.
Upload legacy system documentation (PDF) or point to a GitHub repo — AgentifyX analyzes the codebase, recommends an optimal agent architecture, generates framework-specific boilerplate, and produces enterprise-grade reports with interactive architecture diagrams.

---

✅ Features

Feature	Status
📄 Multimodal PDF extraction (text, tables, images via Gemini Vision)	✅
🧠 Semantic chunking + NER with spaCy	✅
🔍 Vector search with ChromaDB + Sentence Transformers	✅
🤖 Gemini-powered agentic blueprint generation	✅
📊 7-dimension readiness radar chart	✅
🏭 Industry-specific context templates (Healthcare, Finance, etc.)	✅
🔧 Multi-framework boilerplate (CrewAI, LangGraph, AutoGen)	✅
⚠️ HITL checkpoint code generator	✅
🛠️ Automated tools recommendation catalog	✅
💬 Streaming RAG chatbot with session persistence	✅
🕸️ Interactive vis.js agent graph (drag, zoom, click)	✅
🗺️ Mermaid flowchart & sequence diagrams	✅
🔗 GitHub repo analyzer with AST-based code analysis	✅
▶️ Agent pipeline simulation with HITL pause/approve/reject	✅
📄 PDF & DOCX report generation (10 sections each)	✅
📥 Full export (PDF/DOCX/JSON/Code/Diagrams)	✅
🎯 Demo mode (zero API calls, all 7 tabs work)	✅
🩺 Health check endpoint with service status	✅
🌙 Dark theme with premium aesthetics	✅

---

🚀 Quick Start

1. Clone & Setup

git clone https://github.com/Krixna-Kant/AgentifyX.git
cd AgentifyX
python -m venv venv
venv\Scripts\activate        # Windows
# source venv/bin/activate   # macOS/Linux
pip install -r requirements.txt
python -m spacy download en_core_web_sm

2. Configure Environment

Create a .env file in the project root:

GEMINI_API_KEY=your_gemini_api_key_here

Get your key at: Google AI Studio

3. Run

Terminal 1 — FastAPI Backend:

uvicorn app.main:app --reload --port 8000

Terminal 2 — Streamlit Frontend:

streamlit run app/frontend.py

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

4. Demo Mode (For Judges)

Click 🎯 Load Demo in the sidebar to instantly explore all 7 tabs with a pre-computed customer support system analysis — no API calls needed.

---

🏗️ Tech Stack

Layer	Technology
Frontend	Streamlit + Plotly + vis.js
Backend	FastAPI + Uvicorn
AI/LLM	Google Gemini 2.0 Flash
Embeddings	Sentence Transformers (all-MiniLM-L6-v2)
Vector DB	ChromaDB
NLP	spaCy (en_core_web_sm)
Reports	ReportLab (PDF) + python-docx (DOCX)
Database	SQLite (via SQLAlchemy)
Code Analysis	PyGitHub + Python AST

---

📁 Project Structure

AgentifyX/
├── app/
│   ├── main.py                    # FastAPI application
│   ├── frontend.py                # Streamlit 7-tab dashboard
│   ├── schemas/
│   │   └── agent_roster.py        # Pydantic AgentRoster schema
│   ├── services/
│   │   ├── llm_engine.py          # Gemini blueprint generation
│   │   ├── document_parser.py     # PDF extraction
│   │   ├── semantic_pipeline.py   # Chunking + embedding
│   │   ├── boilerplate_generator.py  # Multi-framework code gen
│   │   ├── graph_builder.py       # vis.js graph spec builder
│   │   ├── github_analyzer.py     # GitHub repo + AST analysis
│   │   ├── simulation_generator.py   # Pipeline simulation
│   │   ├── hitl_generator.py      # HITL checkpoint code gen
│   │   ├── tools_matcher.py       # Tools recommendation
│   │   ├── report_generator.py    # PDF report (ReportLab)
│   │   └── docx_generator.py      # DOCX report (python-docx)
│   ├── routers/
│   │   └── chat.py                # Streaming chat endpoint
│   ├── db/
│   │   ├── database.py            # SQLite session storage
│   │   └── models.py              # SQLAlchemy models
│   ├── data/
│   │   ├── industry_templates.json
│   │   └── tools_catalog.json
│   └── demo/
│       └── demo_roster.json       # Pre-computed demo data
├── requirements.txt
├── .env
└── README.md

---

📋 API Endpoints

Method	Endpoint	Description
GET	/api/health	Service health check
POST	/api/v1/process-document	Analyze PDF + optional GitHub
GET	/api/v1/industry-templates	List industry templates
POST	/api/v1/generate-boilerplate	Generate framework code
POST	/api/chat	Streaming RAG chatbot
GET	/api/v1/sessions	List analysis sessions
GET	/api/v1/sessions/{id}	Get session details

---

📜 License

Built for TECHgium 9th Edition hackathon.