The Future of Decision-Making: AI Agent That Predicts Trends & Profits

This is a submission for the AI Agents Challenge powered by n8n and Bright Data

What I Built

I built an AI-powered Intelligence Agent using n8n that transforms raw news into executive-ready reports.

The agent automatically:

Collects breaking news data

Extracts facts, trends, and predictions

Analyzes tone and sentiment

Generates business suggestions (clients, income potential, global expansion strategy)

Highlights convincing points for decision-makers

This helps entrepreneurs, analysts, and executives save time by getting instant insights from complex information.

Demo

SantomaAi

📊 Example Intel Report Output:

📊 Intel Report
━━━━━━━━━━━━━━━━━━━━━━━

📝 Facts

1. New Mexico lawmakers propose legalizing medicinal 'magic mushrooms'.
2. Colorado issues licenses to psychedelic mushroom therapy facilitators.
3. Experts warn of mental health risks due to rising use.

📈 Trend
Legalization of psilocybin across the US and globally.

🔮 Prediction
Mainstream integration of psilocybin therapy in medicine.

🎙️ Tone
Executive briefing.

💡 Business Suggestions

1. Psilocybin Therapy Center
2. Psychedelic Education Services
3. Mushroom-Based Packaging Production

💡 Convincing Points

• Early entry advantage in a booming industry
• Growing consumer demand for natural products
• Educational opportunities in mental health & wellness

n8n Workflow

👉 Santoma Intelligence n8n Workflow

Technical Implementation

• Platform: n8n (self-hosted via Docker)
• Trigger: Webhook node (chat-style interface)
• Data: News / content input
• Processing:
• AI Model (OpenAI GPT) → extracts facts, trends, predictions
• Sentiment analysis → executive tone adjustment
• Structured response formatting with emojis + sections
• Output: Instant Intel Report in Markdown/Chat format

Bright Data Verified Node

Bright Data was used to collect fresh, verified data sources to feed into the AI agent.
This ensures reports are fact-based and up-to-date, not just static prompts.

Journey

Building this agent was both exciting and challenging:

• 🚧 Challenge: Formatting AI output into consistent sections.
• 💡 Solution: I structured the prompts and applied clear headers (Facts, Trend, Prediction, Tone, Suggestions).
• 🚧 Challenge: Webhook was only accessible locally.
• 💡 Solution: Planning to expose with ngrok for testing, and later deploy permanently on a VPS.
• 🚧 Challenge: Making reports business-actionable, not just text.
• 💡 Solution: Added income models, client targets, and expansion strategies in the output.

I learned how powerful n8n + Bright Data + AI can be for building custom intelligence systems without writing full apps.

✨ This project shows how automation + AI can turn scattered data into insightful, ready-to-use intelligence for entrepreneurs and decision-makers.