Online academic content has become a key area of research for text analysis and the development of recommender systems. However, while existing tools typically offer content scraping or text complexity assessment, few integrate both within a unified framework. This paper introduces ASTRA (Academic Scraper for Text and Reading Analysis), a system that automates the collection of scholarly texts and evaluates their linguistic and readability features. ASTRA combines web scraping with Natural Language Processing (NLP) techniques to extract metadata, preprocess text, and compute lexical diversity and readability indexes. Unlike other tools, ASTRA generates text complexity information that supports the development of educational technology, academic writing evaluation, and large-scale corpus studies. Preliminary experiments show that ASTRA effectively retrieves relevant academic material while producing reliable complexity assessments. In general, ASTRA represents a practical resource for researchers and educators seeking to bridge automated data collection with advanced text analysis.