Open-Source Agent Skill Pinpoints Photo Locations Within Meters Using Skylines, Geometry, and Satellite Data

A new open-source agent skill released on GitHub now enables AI coding assistants to geolocate photographs with meter-level accuracy, even when images lack visible landmarks, readable text, or GPS metadata. Called geo-sleuth, the tool transforms Claude Code, Gemini CLI, Cursor, and other AI agents into geolocation investigators by combining multiple data sources: OpenStreetMap geometry, elevation-based skyline matching, sun and shadow angle analysis, satellite imagery, and street-level photography.