Let D = (V (D), A (D)) a digraph. Consider the set PD = {P : P is a non trivial finite directed path in D} and let A and ℬ two subsets of PD. A subset N of V (D) is said to be an (A, ℬ)-kernel of D if (1) for every subset {u, v} of N there exists no uv-directed path P such that P∈A (N is A-independent) and (2) for every vertex x in V (D)∖N there exist y in N and P in ℬ such that P is an xy-directed path (N is ℬ-absorbent). In particular, this is a generalization of the concept kernel by monochromatic directed paths in an m-colored digraph when A=ℬ = {P∈PD : P is a monochromatic directed path}. A classical result in kernel theory establishes that if D is a 2-colored digraph without monochromatic infinite outward paths, then D has a kernel by monochromatic directed paths; this result is known as Sands, Sauer and Woodrow’s theorem. In this paper we show an extension of this result in the context of (A, ℬ)-kernels, for possibly infinite digraphs, as follows: Let D be a digraph, possible infinite, and A and ℬ two subsets of PD such that (A, ℬ)P. Suppose that (1) {V1, V2} is a partition of A such that Vi is Vi-transitive for each i in {1, 2}, (2) for i in {1, 2}, if (xn)n∈N is a Vi-infinite outward path, then there exists j in N such that there exists a directed path from xj+1 to xj in Vi. Then D has an (A, ℬ)-kernel. Generalizations of many previous results are obtained as a direct consequence of this theorem.