End-to-end analytical pipeline for urban air quality monitoring in Bologna using prompting technique

[marcyborg]

Project developed as part of the Short Master in Generative AI — University of Bari Aldo Moro, a.y. 2025/2026. https://github.com/marcyborg/airwatch-bologna

AirWatch Bologna demonstrates how an AI agent can build a complete data pipeline — from discovering a CKAN open data portal to producing an interactive HTML dashboard — using the onData CKAN MCP Server as a programmatic interface for accessing institutional open data. Project developed as part of the Short Master in Generative AI — University of Bari Aldo Moro, a.y. 2025/2026.

Prompting technique: hybrid CoT (Chain-of-Thought) + ReAct (Reason+Act) + Structured Decomposition with final self-check.