Although the answer from @CortAmmon is generally admirable, I think there needs to be a distinction between the type of veracity of the two statements. I would say that:
Statement 1 (bond formation releases energy) is a fundamental truth which emerges from the definition of bond energy.
Statement 2 (anabolic reactions require energy) is a generalization
that is true in most circumstances but is not a fundamental truth, as
illustrated by exceptions.
The first problem is the definition of anabolism. Let us take an uncontraversial one from an internet site:
“Anabolism is… the generation of complex molecule from simple molecules”
or, as in Berg et al. Biochemistry, “Biosynthetic processes”.
It is immediately obvious that this is a much less precise concept than bond energy, and and the distinction arises between anabolism as a whole (e.g. the overall synthesis of glucose from pyruvate in gluconeogenesis) and individual anabolic reactions (e.g. the aldolase reaction in gluconeogenesis).
Now in anabolism in biology we are generally concerned with the the creation of complex molecules containing new carbon–carbon bonds, peptide bonds, glycosidic bonds, phospho-diester bonds and the like. As @CortAmmon makes clear, the second statement is valid because (I would say “if”) the bonds in the molecules that have to be broken for the likage to occur have a greater total bond energy than that of the new bond in the product. This depends on the particular chemistries involved, and may turn out to be true for the whole process, but there seems to be no fundamental reason why it should be so.
Certainly, it need not be true for individual anabolic reactions, as illustrated by the aldolase reaction, mentioned above:
This is a reversible reaction, which in the anabolic direction forms a C–C bond between two trioses. The standard free energy change in this reaction is near zero.
If the poster’s textbook had made this distinction, the poster’s confusion on the matter need not have arisen.