Large Language Models (LLMs) excel at synthesising globally documented knowledge but lack the fine-grained, real-time awareness required for field-level agricultural and rural-planning decisions. This paper introduces JackDaw, a spatially enabled chat-agent architecture that couples foundation-model reasoning with multi-modal geospatial data streams and a retrieval-augmented generation (RAG) pipeline. JackDaw implements a tool-prefiltering mechanism that selects only those data connectors whose topical, temporal and spatial metadata match the current query, thereby mitigating the diminishing returns observed when LLMs are exposed to large, flat toolsets. Through LangChain-based orchestration the platform dynamically assembles workflows that range from lightweight natural-language processing models to domain-specific analytic kernels, while a value-engineering strategy allocates computationally intensive models (e.g., GPT-4-class) only to tasks that require broad contextual reasoning. Benchmark experiments on forestry-asset discovery and vineyard-site assessment demonstrate that JackDaw delivers location-specific, traceable answers that outperform a standalone proprietary LLM, which provides only generic or spatially misattributed responses. The results confirm that bridging global language models with local spatial intelligence markedly reduces hallucination rates and enhances the operational readiness of AI for sustainable agriculture and rural development. Index Terms—Large language models; geospatial AI; retrieval-augmented generation; context-aware agriculture; spatial decision support; tool prefiltering; JackDaw system; rural planning.
Abstract: This study explores a methodology for assessing territorial innovation potential using OpenStreetMap (OSM) data and geoinformation technologies. Traditional assessment methods often rely on aggregated statistical data, which provide a generalized view but overlook the spatial heterogeneity within regions. To address this limitation, the proposed methodology utilizes open, up-to-date OSM data to identify key infrastructure elements, such as universities, research institutions, and data centers, which drive regional innova- tion. The methodology includes data extraction, harmonization, and spatial analysis using tools like QGIS and kernel density estimation. Results from the PoliRuralPlus project pilot regions highlight significant differences in innovation potential between urban centers and rural areas, emphasizing the importance of detailed spatial data in policy making and regional development planning. The study concludes that OSM-based assessments provide spatially detailed targeted, flexible, and replicable insights into regional innovation potential compared to traditional methods. However, the limitations of crowdsourced data, such as variability in quality and completeness, are acknowledged. Future devel- opments aim to integrate OSM with official statistical data and other data resources to support more efficient and fair resource allocation and strategic investments in regional innovation ecosystems.