Plants develop various biochemical, physiological and molecular mechanisms to sense a mixture of stress signals and elicit a precise response to minimize the damages. Therefore, in-depth studies are required to understand the genetic bases behind the plant’s tolerance in response to environmental stresses. Agave, monocotyledonous succulent plant, is endemic to arid regions of North America, exhibiting exceptional tolerance to their xeric environments. Genomic resources of Agave species have received little attention irrespective of their cultural, economic and ecological importance, which so far prevented the understanding of the molecular basis underlying their adaptations to the arid environment. To elucidate the drought-responsive mechanisms, here RNA-Seq libraries derived from the Agave sisalana leaves under control and drought conditions have been prepared and sequenced. More than 278 million paired ends Illumina leaf specific reads were generated. A Comparative de novo approach was applied to assemble paired-end reads into 93,141 contigs and 67,328 unigenes. Blast analysis of these unigenes against the non-redundant public databases (nr, swiss_prot, interProScan, Pfam, Viridi_plante, Pfam, Plant_TF, GO, KEGG and COG resulted in 37,546 unigenes with gene descriptions, functional categorization, or gene ontology terms. The expression study unveiled 3,095 differentially expressed unigenes between well-irrigated and drought-stressed leaf samples. Gene ontology and pathway analysis specified a significant number of abiotic stress responsive genes and pathways involved in processes like hormonal responses, antioxidant activity, and response to stress stimuli, wax biosynthesis, and ROS metabolism. Transcripts to several families belonging harboring important drought-response were also reported. Furthermore, Insilico 36,525 high confidence variants position (SNPs), 13,375 microsatellite markers (SSR) are detected in the annotated unigenes and 8164 marker specific pair of primers were designed. Stable internal housekeeping genes’ identification was carried out for accurate normalization of the target gene expression by qRT-PCR in Agave sisalana. In total 15 candidate’s housekeeping genes from de novo assembled transcriptome data were screened out for further evaluation. These includes ADP-ribosylation factor 2 (ARF2), Cyclophilin A (CYCA), Ribulose Bisphosphate Carboxylase activase B (RcaB), Rubisco Activase (RCA), Actin 11 (ACT11), beta-tubulin 4 (β-Tub 4), Eukaryotic elongation factor 1-alpha (EEF1α), eukaryotic initiation factor-4A (eIF-4A), Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), polyubiquitin (UB), RNA polymerase II (RPII), RuBisCO small subunit (RBCs), Serine/Threonine-protein phosphatase catalytic subunit (PP2A-1), Cullin-1 (CUL-1), WIN1, Ubiquitin 10 (UB10) and Ubiquitin-Conjugating enzyme (UBE2). The expression stability of these reference genes was rigorously analyzed and ranked in order by using four different statistical algorithms; NormFinder, BestKeeper, geNorm, and RefFinder under drought, rehydration, heat (± 60 °C), cold (± 4°C) and salt stress (100mM to 400mM) conditions. β-Tub 4, PP2A-1 and β-Tub 4, ARF2 were the most stable reference genes under drought and rehydration condition respectively. To heat stress (high-temperature), CYCA and GAPDH were the stable reference genes while CUL-1 and WIN1 were the most stable reference genes under cold stress condition. For Salt stress, β- Tub 4 and RP II was the most appropriate leaf specific housekeeping genes in Agave sisalana. To validate the ranking of reference genes, a qRT-PCR assay of AsHSP20 as target gene was conducted by using the most suitable and least reliable reference genes under abiotic stress condition. Relative absolute quantification of the target AsHSP20 gene was carried out to determine the copy number under different abiotic stress and rehydration condition, which further confirmed the reliability of studied reference genes. Taken together this study ranked the reference genes from most to least reliable order for counts data normalization. This suggests that the use of appropriate reference genes is critical for gene expression studies under specific conditions. Drought specific stable β-Tub 4 gene was used as an internal control to validate the differentially expressed genes expression data using the quantitative real-time polymerase chain reaction. This study presents the first insight into the genomic structure of A. sisalana underlying adaptations to drought stress, which not only provided a rich genomic resource for gene discovery and marker development but will also facilitate further to understand the complexity underlying drought tolerance and adaptation in agave and other plant species.