The last component of the control knowledge consists of two sets of meta-rules, that are responsible for tuning the inference at the object-level, or of implementing it, in the case of a reified evaluation strategy, and of managing the hierarchical structure of submodules by adding or cancelling dependencies. Meta-rules in both sets have a similar syntactical structure, that is, a conjunction of meta-predicates as premise and a set of meta-predicates as conclusion. The only difference is on the meta-predicates allowed as conclusion in each set. In this section we describe the behaviour of the predefined meta-predicates used in the premises and in the conclusions. Apart from them, any user-defined meta-predicate can be used in the premises or in the conclusions. Negation in the conditions is defined as negation by failure.