Gyrochronology is the technique that uses rotation periods to derive stellar ages via age-rotation relations. If properly calibrated, its potential applications to Galactic, stellar and exoplanetary astrophysics are far-reaching. With the continuous increase of the number of stars with measured rotation periods, it has become imperative to comprehensively test gyrochronology and understand its limits. In this talk I will present a novel method that tests the modern age-rotation relations in under-explored domains using wide binary stars. The components of a given binary are co-eval and taken together they can provide exquisite gyrochronology assessments. By applying this method to a sample of Kepler-field binaries and star clusters we find that: 1) the commonly used age-rotation relations do have predictive power in identifying co-eval populations of field stars; 2) they achieve the best results when used in young clusters ($<$ 1 Gyr); 3) their performance decreases when used in field wide binaries and older clusters. All in all, better calibrations are needed, particularly in the regime of old ages ($>$ few Gyr). This work sets up the stage for decisive gyrochronology assessments in the era of ever-expanding space-based photometry.