This article reconstructs the Proto-Kampa (Arawakan: Peru) consonant inventory using lexical data from Nomatsigenga, Ashéninka, Pajonal Ashéninka, Asháninka, Kakinte, and Matsigenka. Cognate set and correspondence set construction was partly automated by the use of LingPy, Edictor, and Lingrex—computational tools developed by Johann Mattis-List. I agree with Michael (2010) in reconstructing an inventory of /m, n, N, p, b, t, k, g, ts, tʃ, s, ʃ, h, ɾ, j/. Sound changes that affected these Proto-Kampa consonants in the daughter varieties include palatalization, the development of contrastive aspiration in Ashéninka and Pajonal, *Np > m and *Nk > ŋ in Nomatsigenga, lenition and loss of *g, and loss of *h.