Technical documentation and training materials are important elements in helping to accelerate the use and impact of agricultural research for development. The creation and delivery of these resources can be enhanced through content enrichment and the production and reuse of modular components. This process can be further improved by integrating rich semantic descriptions with resource metadata and domain-specific markup combined with the consistent use of controlled vocabulary. The Darwin Information Typing Architecture (DITA) supports both the integration of metadata as well as markup for technical as well as learning and training content. DITA also includes mechanisms for adding the semantics of taxonomy and ontology definitions for classifying content. This paper explores the potential use of enriching agroforestry learning resources with DITA markup through descriptive metadata and subject classification.