🚀 AI Task Analyst Web

A smart AI-powered web app for analyzing and managing project/task data, featuring advanced function calling and data analysis. Enjoy an AI chat that remembers previous conversations and delivers actionable insights.

---

📊 Flow Diagram Example

---

🌟 Key Features

• 🤖 Intelligent AI Chat with session-based memory
• 🔍 Function Calling for precise data search and filtering
• 📈 Advanced Data Analysis with statistics and insights
• 💬 Automatic Conversation Routing based on question type
• 📱 Responsive Frontend built with Vue.js
• 📝 Comprehensive Logging for monitoring and debugging
• 🎯 Optimized Prompt Engineering for best results

---

🧠 AI Capabilities

1. Data Search & Filtering

• Search tasks by user/employee name
• Filter by status (todo, in-progress, done)
• Search by project/task name
• Filter by specific month

2. Complex Analysis

• Team and individual performance statistics
• Productivity comparison between users
• Monthly work trend analysis
• Identify bottlenecks and improvement areas
• Executive reports with actionable insights

3. User Management

• Complete team member list
• Detailed user and task assignment info

4. Conversational AI

• Natural conversation in English (and Indonesian)
• Explains system features and capabilities
• General help and usage guidance

---

🗂️ Project Structure

ai-task-analyst/
│
├── app.py                    # FastAPI backend with function calling
├── requirements.txt          # Python dependencies
├── .env                      # LLM API key
├── database_task.sql         # SQL schema + sample data
├── static/
│   └── index.html            # Responsive chat frontend
├── tasks.db                  # (auto-generated) SQLite DB
├── app.log                   # Log file for monitoring
├── Dockerfile                # Docker image configuration
├── docker-compose.yml        # Production Docker Compose
├── docker-compose.dev.yml    # Development Docker Compose
├── nginx.conf                # Nginx reverse proxy config
├── .dockerignore             # Docker build ignore file
└── README-Docker.md          # Docker documentation

---

🏗️ System Architecture

🤖 Routing Intelligence

The system uses a smart AI router to select the right function:

1. search_activities - Search and filter data
2. get_user_list - List users/employees
3. analyze_full_data - Complex analysis and statistics
4. general_conversation - General chat

🔗 Function Calling

• search_activities(user_name, status, task_name, month)
• get_user_list()
• Context stuffing for complex analysis
• General conversation for common questions

---

⚡ Installation & Setup

🐳 Option A: Docker Installation (Recommended)

Prerequisites

• Docker Desktop atau Docker Engine
• Docker Compose

Quick Start with Docker

# 1. Clone the repository
git clone <repo-url>
cd ai-task-analyst

# 2. Create .env file
cp env.example .env
# Edit .env with your API keys

# 3. Run with Docker (Development)
docker-compose -f docker-compose.dev.yml up --build

# 4. Or run with Docker (Production)
docker-compose up --build

# 5. Access the application
# Open: http://localhost:8000/static/index.html

Docker Commands

# Development mode (with hot reload)
docker-compose -f docker-compose.dev.yml up --build

# Production mode
docker-compose up --build

# With nginx reverse proxy
docker-compose --profile production up --build

# Stop containers
docker-compose down

# View logs
docker-compose logs -f

Docker Features

• ✅ Hot Reload untuk development
• ✅ Nginx Reverse Proxy untuk production
• ✅ Health Checks untuk monitoring
• ✅ Volume Persistence untuk database dan logs
• ✅ Environment Variables support
• ✅ Network Isolation untuk keamanan

📖 Docker Documentation: Lihat README-Docker.md untuk panduan lengkap Docker.

---

🐍 Option B: Manual Python Installation

1. Clone the repo & enter the folder

git clone <repo-url>
cd ai-task-analyst

2. Create & activate environment (optional but recommended)

# Using conda
conda create -n ai-task python=3.10
conda activate ai-task

# Or with venv
python -m venv venv
venv\Scripts\activate  # Windows
source venv/bin/activate  # Linux/Mac

3. Install dependencies

pip install -r requirements.txt

4. Prepare the .env file

Create a .env file in the root directory:

LLM_API_KEY=your_google_ai_studio_api_key
LLM_API_URL=https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent?key=your_google_ai_studio_api_key
OLLAMA_HOST=http://127.0.0.1:11434
OLLAMA_MODEL=llama2

Replace with your actual API key and endpoint as needed.

5. Ensure database_task.sql contains CREATE TABLE + INSERT

Already included in the repo, but make sure the top of the file looks like:

CREATE TABLE IF NOT EXISTS user (...);
CREATE TABLE IF NOT EXISTS task (...);
CREATE TABLE IF NOT EXISTS activity (...);
-- then INSERT ...

---

🚦 Running the Application

🐳 Docker Method (Recommended)

# Development mode (with hot reload)
docker-compose -f docker-compose.dev.yml up --build

# Production mode
docker-compose up --build

# With nginx reverse proxy
docker-compose --profile production up --build

🐍 Manual Python Method

# Start the backend
uvicorn app:app --reload

• The server will auto-create tasks.db from database_task.sql if it doesn't exist.
• Logs are written to app.log for monitoring.

🌐 Access the Application

Open in your browser:

• Docker: http://localhost:8000/static/index.html
• Manual: http://127.0.0.1:8000/static/index.html
• With Nginx: http://localhost/static/index.html

---

💡 Usage Examples

Data Search

• "Find tasks assigned to Budi"
• "Show tasks with status done"
• "Activities in January"
• "Website project tasks"

Complex Analysis

• "Who is the most productive?"
• "How many tasks were completed this month?"
• "Compare team performance"
• "Generate monthly report"
• "Work statistics"

User Management

• "Who are the team members?"
• "Show all employees"
• "User list"

General Conversation

• "Hello, how are you?"
• "What can you do?"
• "Thank you"

---

🛠️ Troubleshooting & Error Handling

🐳 Docker Issues

Error: Port already in use

# Check what's using port 8000
netstat -tulpn | grep :8000

# Kill process using port 8000
sudo kill -9 $(lsof -t -i:8000)

# Or change port in docker-compose.yml
ports:
  - "8001:8000"  # Use port 8001 instead

Error: Container won't start

# Check container logs
docker-compose logs -f

# Rebuild without cache
docker-compose build --no-cache

# Remove and recreate containers
docker-compose down -v --remove-orphans
docker-compose up --build

Error: Database issues in Docker

# Remove existing database
rm tasks.db

# Restart container (will recreate database)
docker-compose restart

🐍 Manual Python Issues

Error: sqlite3.OperationalError: no such table: user

Cause: The tasks.db file exists but is empty or has the wrong schema.

Solution:

1. Delete the tasks.db file in the project folder.
2. Restart the server (uvicorn app:app --reload).
   • The server will automatically recreate the database from database_task.sql.

Error: Import "fastapi" could not be resolved

• Make sure your environment is active.
• Run: pip install fastapi uvicorn

Error: uvicorn not found

• Run: pip install uvicorn

🔧 General Issues

Error: LLM not responding/API error

• Ensure .env is correct and API key is active.
• Check app.log for error details.
• Make sure the API response format matches expectations.

Error: 404 Not Found for endpoint

• Use the /chat endpoint (not /chat_stream).
• Make sure the frontend is updated to use the correct endpoint.

Error: Permission denied (Docker)

# Fix file permissions
sudo chown -R $USER:$USER .

# Or run with user permissions
docker-compose run --user $(id -u):$(id -g) ai-task-analyst

---

🧑‍💻 Technical Features

Enhanced Prompts

• Router Prompt: Smart AI for function selection
• Summary Prompt: Informative and structured data presentation
• Analysis Prompt: Deep analysis with actionable insights
• General Prompt: Friendly, natural conversation

Response Format

• HTML tables for structured data
• Markdown support for formatting
• Code blocks for technical output
• Responsive design for mobile

Performance

• 45-second timeout for API calls
• Detailed logging for monitoring
• Robust error handling
• Efficient session management

---

📚 Usage Notes

• All data is sourced from the SQLite database (tasks.db).
• To update data, edit database_task.sql, delete tasks.db, and restart the server.
• Chat history per session is stored in the database.
• Detailed logs are available in app.log for debugging.

---

🚀 Further Development

Docker Deployment

• For production, use docker-compose --profile production up -d
• Add SSL/HTTPS certificates to nginx configuration
• Use Docker Swarm or Kubernetes for scaling
• Implement health checks and monitoring

Manual Deployment

• For large datasets, limit data sent to the LLM.
• Switch LLM to OpenAI, Ollama, etc. by updating .env.
• For deployment, use a production server (e.g., uvicorn app:app --host 0.0.0.0 --port 80).
• Add authentication and authorization as needed.
• Implement streaming responses for better UX.

General Improvements

• Add more AI models support (OpenAI, Anthropic, etc.)
• Implement real-time notifications
• Add user authentication and role-based access
• Create mobile app version
• Add data export/import features

---

📝 License

Free to use for learning and development purposes.

---

If you encounter other errors, check app.log or ask here!