Finally got around to getting kcst in publishable shape.

kcst predicts species and MLST for bacterial genome assemblies, and does it fast: it takes less than a second for the average assembly. kcst comes with khc (k-mer hit counter), a fast and accurate tool to compute sequence similarity between a query sequence and a set of subject sequences.

Homepage is here: http://io.zwets.it/kcst, and here is the code.