Proceedings of the International scientific and practical conference ―Science in the Era of Globalization‖ (May 22-24, 2026) / Publisher website: www.naukainfo.com. - Zurich, Switzerland, 2026. - 353 p.

118 The rapid development of financial technologies has significantly increased the importance of comprehensive and high-quality financial data. Modern financial analytics increasingly relies on hybrid approaches that combine dynamic data with contextual information, enabling more accurate modeling of financial offers and market behavior. However, such analysis requires access to diverse and heterogeneous data sources, which are often fragmented and inconsistent. Financial information extends far beyond transactional data and includes a wide range of internal and external sources that must be consolidated to improve analytical outcomes. In particular, the work [1] highlights the necessity of combining traditional financial data with alternative and external datasets to improve decision-making and modeling of financial offers. Similarly, contemporary research on financial data architectures demonstrates that modern financial systems operate in highly heterogeneous environments, integrating structured and unstructured data through scalable pipelines and cloud-based infrastructures [2]. Ukrainian domestic government bond (OVDP) market represents a relevant case study of a data-intensive financial environment. Data related to OVDP instruments are provided by multiple entities, including banks, brokers, and public institutions, each offering different subsets of information. In addition, some institutions enable bond trading operations without providing structured or programmatically accessible data, which complicates automated data collection and integration. This paper addresses the scientific and applied problem of collecting, consolidating, and normalizing data on Ukrainian domestic government bonds from heterogeneous sources. Each bond transaction is considered as a financial offer characterized by a set of parameters, including yield, maturity, currency, price, as well as contextual market conditions. The study focuses on practical approaches to data acquisition from financial institutions and brokerage platforms, taking into account issues such as missing data, inconsistent formats, and limited accessibility. The main contribution of this work is the development of a data collection framework supported by an MCP-based server that orchestrates data retrieval,