Large language models (LLMs) are becoming more and more important for the revolutionary transformation in the world, bringing new ideas to the intelligent development of the aviation field. It is necessary to conduct research on the construction of national defense aviation intelligent open source systems based on large language models, which can improve the efficiency of information acquisition and intelligent analysis capabilities. This study takes the application of an aviation open source intelligence recommendation system based on LLMs as an example, which automatically collects and classifies information through the integration of LLMs. It is able to build a local database and information model, and realize rapid intelligence integration and intelligent analysis capabilities. LangChain architecture is used to carry out large language models interaction and prompt on external open source data to enhance the accuracy of results, aiming to provide basic application services for the intelligent development of the aviation field.