Shear strength parameters, including the internal friction angle φ and cohesion c, are critical for the design and construction of foundations and underground excavations. The measurement of φ and c through laboratory tests (e.g., triaxial tests and direct shear tests) and conventional in-situ tests (e.g., field shear tests) is time-consuming and costly. Therefore, accurate indirect estimation of φ and c via convenient and economical approaches holds significant practical value. This study proposes an efficient approach to estimate φ and c using three portable test-derived indices: the point load strength index I_s(50), the Schmidt hammer rebound value S_rn, and the ultrasonic S-wave pulse velocity V_s. A total of 102 groups of rock specimens (quartzite, limestone, and marble) from Malaysia were tested to develop 42 estimation models using single-, dual-, or tri-index as input variables, while an additional 20 groups were employed for model validation. The fitting performance and prediction accuracy of these models were evaluated using the coefficient of determination (R²) and relative error percentage. The results show that tri-index-based models (I_s(50) + S_rn + V_s) achieve the highest precision (R² = 0.86−0.91, REP = 2.0−7.7 %), followed by dual-index models (I_s(50) + S_rn and I_s(50) + V_s) and single-index models. Compared with previous studies employing artificial intelligence-based models, this study applied traditional statistical approaches that are transparent, easy to implement, and do not require complex computational tools. Although the achieved precision is moderate compared with advanced machine learning techniques, the results remain reliable and practical, given the variability of rock types and geological conditions. For engineering applications, tri-index models are recommended for high-precision requirements, while I_s(50)-based single- or dual-index models are suitable for general field use. This research provides simple, indirect tools for estimating the shear strength of rock materials.