[en] Clouds and related atmospheric phenomena have a big influence on the quality of astronomical observations. Especially in case of ground-based gamma telescopes such as FACT, bad atmospheric conditions impair the reconstruction of air-shower events. The First G-APD Cherenkov Telescope aims for automatic long-term monitoring. Therefore, it benefits greatly from an advanced scheduling algorithm which takes into account the current weather conditions. While there is no way to reduce the occurrence of clouds in the direction of a desired object, it is possible to increase the duty cycle by switching to an uncovered source. Hence, a quantitative rating of the sky cloudiness is needed to differentiate between covered and uncovered areas. This talk presents a method to calculate the current sky cloudiness by searching stars in 180 allsky camera images. This method can be applied to arbitrary areas, e.g., the whole sky or a few degrees around any certain source. Results and various visualisations are presented, such as the distribution of the parameters for different weather conditions. Moreover, their developement over time is shown for multiple sources in a partly clouded night. And a first approach for predicting the cloud's movement by using subsequent images and additional data such as wind profiles is discussed.