Abstract Introduction Pyrazinamide, one of the first-line antituberculosis drugs, displays variability in drug exposure that is associated with treatment response. A simple, low-cost assay may be helpful to optimize treatment. This study aimed to develop and validate a point-of-care assay to quantify the concentration of pyrazinamide in saliva. Methods All measurements were conducted using the nano-volume drop function on the mobile ultraviolet (UV) spectrophotometer (NP80, Implen, Germany). Assay development involved applying second derivative spectroscopy in combination with the Savitzky–Golay filter between wavelengths of 200–300 nm to increase spectral resolution. Assay validation included assessing selectivity, linearity, accuracy, precision, carry-over and matrix effects. Specificity was also analysed by evaluating the impact of co-administered medications on pyrazinamide results. Sample stability was measured at various temperatures up to 40°C. Results The calibration curve (7.5–200 mg/L) was linear (R2 = 0.9991). The overall accuracy (bias%) and precision (CV%) ranged from −0.66% to 5.15%, and 0.56% to 4.95%, respectively. Carry-over and matrix effects were both acceptable with a bias% of <±4% and CV% of <7.5%. Commonly co-administered medications displayed negligible interferences. Levofloxacin displayed analytical interference (bias% = −10.21%) at pyrazinamide concentrations < 25 mg/L, but this will have little clinical implications. Pyrazinamide was considered stable in saliva after 7 days in all storage conditions with a CV% of <6.5% and bias% of <±10.5% for both low- and high-quality control concentrations. Conclusions A saliva-based assay for pyrazinamide has been successfully developed and validated using the mobile UV spectrophotometer.