Constructing The Dataset Of Iran's Missile War Against Israel
Published: 23 - Mar - 2026
Since Iran began its direct missile and drone attacks against Israel in April 2024, I've been building an open-source intelligence (OSINT) dataset to document every aspect of these unprecedented strikes. The project is called Promise Denied, and it tracks the ballistic missiles, cruise missiles, and drones that Iran has fired at Israel and coalition targets across multiple rounds of attacks.
Why Build This Dataset?
The scale of Iran's attacks against Israel is historically significant — these represent the first direct state-on-state ballistic missile engagements of their kind. Yet the data is scattered across news reports, defense analyses, and social media. I wanted to bring it all together into structured, machine-readable datasets that researchers, journalists, and analysts could use freely.
What The Dataset Tracks
The project documents four distinct attack rounds with structured data covering:
Munitions fired — types, quantities, and technical specifications of each missile and drone
Target locations — mapped geographically across 13+ countries
Defense systems engaged — Arrow-3, THAAD, Iron Dome, Patriot PAC-3, David's Sling, and coalition assets
Interception rates and casualty figures for each round
Temporal data — launch times and salvo sequences
Open Data, Open Source
All of the data is freely available. You can access the full datasets on GitHub, Kaggle, and Hugging Face. The project welcomes community corrections and contributions — the goal is accuracy and transparency.
Data is sourced from news reporting and open-source intelligence, with explicit caveats that timestamps are approximate and munitions counts may vary across sources. This transparency is intentional — I'd rather show the uncertainty than present false precision.
Built With AI Assistance
This project was made possible in large part with Anthropic's Claude, which helped with research, data structuring, and analysis. It's a good example of how AI tools can augment open-source research — helping to process large volumes of reports and structure them into consistent, queryable formats.
Visit the project at promisedenied.com.
Daniel Rosehill
AI developer and technologist specializing in AI systems, workflow orchestration, and automation. Specific interests include agentic AI, workflows, MCP, STT and ASR, and multimodal AI.
