Spatial transcriptomics enables researchers to study gene expression within tumor microenvironments at single-cell resolution while maintaining tissue architecture. However, analyzing spatial omics data is challenging due to the complexity of outputs, including large-scale images, millions of transcripts, and numerous spatial coordinates. Moreover, standard volumetric segmentation is often hampered by optical blurring and signal bleed-through in out-of-focus z-planes, while built-in tools in platforms like Xenium are typically limited to 2D analysis. To overcome these challenges, we introduce STAT, a transparent, time-efficient, and customizable pipeline for analyzing whole-tumor biopsy slides. STAT addresses the limitations of full-volume 3D approaches by implementing an 'auto-focus' strategy that selectively integrates only the sharpest optical planes, maximizing the signal-to-noise ratio while retaining 3D cellular integrity. By integrating standard community tools (e.g., Seurat), the pipeline facilitates accurate multi-plane cellular segmentation, cell-to-transcript mapping, and comprehensive images' z-stack analysis. This versatile tool streamlines critical steps in spatial omics analysis, enabling high-fidelity single-cell spatial resolution and enhancing the exploration of tumor microenvironment. The pipeline is structured in a Docker container which encloses several programming languages such as Python and R , together with the Image analysis software QuPath. STAT pipeline runs through the computational workflow Snakemake. The analysis is based on large OME.TIFF files generated with the in-situ spatial transcriptomics platform named Xenium (by 10x Genomics). The computational heaviness of the images is successfully overcame by STAT pipeline through its image-size reduction logic, derived from its graph-based feature selection machine learning approach. The image size reduction is the base for the next computational steps of the pipeline which include 2D or 3D nuceli segmentation with Cellpose software, followed by tailored cell-to-transcripts assignation and single cells automatic annotation. Eventually STAT pipeline generates a standard RDS format file compatible with R studio together with informative quality control plots and dimensionality reduction graphs displaying cell clusters distribution. STAT pipeline demonstrated computational superiority and higher cellular sensitivity when benchmarked against current lead transcriptomics pipelines such as Sopa and Baysor. Now, with the automatic decision-making process of STAT , users with different degree of experience in bioinformatics or biology will be enabled to conduct image-tailored spatial transcriptomics analysis at sub-cellular level.