OpsPilot - Multi-Agent Workplace Automation System

Intelligent Workplace Automation powered by Multi-Agent AI

---

A comprehensive AI-powered workplace automation platform leveraging multiple specialized agents orchestrated through LangGraph, with intelligent memory, caching, and governance systems.

🎯 Overview

OpsPilot is an intelligent workplace assistant that automates email triage, meeting management, task prioritization, follow-up tracking, wellness monitoring, and report generation using a multi-agent architecture with ReAct pattern implementation.

Key Features:

• 🤖 6 Specialized Agents - Email, Meeting, Tasks, Follow-up, Wellness, Reporting
• 🔄 LangGraph Orchestration - Dynamic agent routing and workflow management
• 🧠 Vector Memory - Semantic search and learning from past interactions
• 🛡️ Governance Layer - Policy enforcement, approvals, audit logging
• 📊 Real-time Analytics - Wellness scoring, productivity metrics, cost tracking
• 🎨 Modern UI - Next.js dashboard with real-time updates

📁 Project Structure

OpsPilot/
├── agents/                 # Specialized AI agents (ReAct pattern)
├── orchestration/          # LangGraph workflows & routing
├── backend/                # FastAPI server & API routes
├── frontend/               # Next.js web application
├── memory/                 # Vector store & episodic memory
├── governance/             # Policy, audit & cost management
├── repos/                  # Data access layer
├── config/                 # Configuration management
├── tests/                  # Test suites
├── chroma_db/              # Vector database persistence
└── episodes/               # Recorded agent episodes

🚀 Quick Start

Prerequisites

• Python 3.10+
• Node.js 18+
• Git

Backend Setup

cd backend
pip install -r requirements.txt
python app.py

Backend runs on http://localhost:8002

Frontend Setup

cd frontend
npm install
npm run dev

Frontend runs on http://localhost:3000

🤖 Agents

Agent	Purpose	Key Functions
Email Agent	Email triage & summarization	summarize(), extract_actions(), draft_reply()
Meeting Agent	Meeting management & MoM	generate_mom(), extract_decisions()
Tasks Agent	Task prioritization & tracking	prioritize(), extract_actions()
Follow-up Agent	Follow-up tracking & reminders	check_followups(), create_nudge()
Wellness Agent	Wellness scoring & recommendations	calculate_score(), suggest_breaks()
Reporting Agent	EOD/Weekly reports	generate_eod_report(), generate_weekly_report()

📊 Technology Stack

Backend

• Framework: FastAPI
• Orchestration: LangGraph
• LLM: Azure OpenAI / Claude
• Memory: ChromaDB + Custom Vector Store
• Database: JSON (mock data)
• Cache: LLM Response Cache

Frontend

• Framework: Next.js + React
• Styling: Tailwind CSS
• Language: TypeScript
• UI Components: Shadcn/ui

Infrastructure

• Governance: Policy enforcement, Rate limiting
• Audit: Complete audit trail logging
• Monitoring: LLM usage & cost tracking

🏗️ Architecture

System Architecture Diagram

High-level system architecture showing agent orchestration and data flow

Workflow Visualization

LangGraph orchestration showing agent routing and decision flow

Data Flow

User Request
    ↓
[Super-Graph] → Intent Classification
    ↓
[Specialized Agent] → Processing (ReAct pattern)
    ↓
[Memory Layer] → Vector store + Episodic memory
    ↓
[Governance] → Approval gates + Audit logging
    ↓
API Response → UI Update

State Management

• Workflow State: LangGraph
• Agent Memory: Episode storage (ChromaDB)
• Vector Memory: Semantic search & retrieval
• API State: Backend persistence

🔐 Security & Governance

• Approvals: Sensitive actions require human approval
• Audit Trail: Complete operation logging
• Cost Management: LLM token & cost tracking
• Rate Limiting: API and LLM rate limits
• Policy Engine: Configurable governance rules

📖 Documentation

• ARCHITECTURE.md - Full system design & diagrams
• CODEBASE_INDEX.md - Complete module reference
• frontend/README.md - Frontend setup guide
• backend/README.md - Backend setup guide

🧪 Testing

# Run backend tests
cd backend
python -m pytest

# Run frontend tests
cd frontend
npm test

📊 Configuration

Configuration is centralized in config/settings.py:

SETTINGS = {
    "env": "dev",          # dev, staging, prod
    "data": { ... },       # Data paths
    "governance": { ... }, # Policy settings
    "agents": { ... }      # Agent configuration
}

Use .env.example as a template for your environment setup.

🔄 Workflows

Email Processing

1. Monitor inbox
2. Summarize & extract actions
3. Draft replies
4. Governance approval (if needed)
5. Store in memory for future reference

Meeting Management

1. Schedule meeting
2. Generate minutes (MoM)
3. Extract decisions
4. Track follow-ups
5. Archive for future reference

Task Management

1. Receive task request
2. Prioritize against workload
3. Assign resources
4. Track progress
5. Report completion

🚦 Environment Variables

Copy .env.example to .env and configure:

# LLM Configuration
AZURE_OPENAI_KEY=...
AZURE_OPENAI_ENDPOINT=...
AZURE_OPENAI_MODEL=...

# Data Configuration
DATA_DIR=./data/mock_data_json

# Server Configuration
API_PORT=8002
API_HOST=0.0.0.0

# Governance
ENABLE_APPROVALS=true
COST_LIMIT_USD=100

📈 Monitoring

Track agent performance and costs:

python print_graph.py    # View workflow graph
python show_tasks.py     # Display active tasks

🤝 Contributing

1. Create a feature branch
2. Make your changes
3. Test thoroughly
4. Submit a pull request

📧 Support

For questions or issues:

• 📖 Check CODEBASE_INDEX.md
• 🐛 Open an issue
• 💬 Start a discussion

---

Last Updated: February 5, 2026
Version: 1.0.0