Medulloblastoma is a tumor of the cerebellum and is the most common malignant brain tumor of childhood. Current therapies, while effective, introduce severe negative consequences to patients, and new molecular approaches are needed. Overactivated sonic hedgehog (SHH) signaling causes about 30% of medulloblastomas. In vertebrates, SHH signaling requires the primary cilium, and pathway members – including SHH’s receptor patched1 (Ptch1), obligate pathway transducer smoothened (Smo), and Gli family transcription factors – dynamically traffic in and out of cilia. Mutations in SHH pathway genes (including Ptch1 and Smo) cause medulloblastoma, but Gli repressor has been shown to be protective against medulloblastoma formation. We study a ciliary GTPase called Arl13b that uniquely regulates SHH, both at the level of SMO and downstream of SMO, where it regulates Gli activator but not Gli repressor. Loss of Arl13b results in ligand-independent constitutive low-level pathway activation but prevents maximal signaling and spares Gli repressor, making ARL13B an attractive target for developing therapies. We hypothesized that the loss of Arl13b could reduce the high levels of pathway activation responsible for SHH-derived medulloblastoma formation while leaving Gli repressor intact, without ablating pathway activity completely. Here, we use a conditional allele of Arl13b in the mouse to show that ARL13B plays a role in cerebellar development through the regulation of proliferation and that its loss dramatically impacts SHH pathway output in a variety of conditions. Deletion of Arl13b effectively reduces signaling levels in the presence of constitutive pathway activation driven by an oncogenic form of SMO, and knockdown of ARL13B in human medulloblastoma cell lines reduces their clonogenic potential. Taken together, our in vivo and in vitro studies demonstrate that ARL13B is a novel target for the treatment of medulloblastoma.