The homomorphisms (14), (15), (16) are morphisms of complexes.
Let us prove it for example for the homomorphism (14). Denote
κ:HomgrA(C′,C′′)⊗kHomgrA(C,C′)→HomgrA(C,C′)
this homomorphism. Let f∈HomgrA(C′,C′′)p and g∈HomgrA(C,C′)q; then by definition one has κ(f⊗g)=f∘g. Moreover :
D(f⊗g)=Df⊗g+(−1)pf⊗Dg=(d′′∘f)⊗g−(−1)p(f∘d′)⊗g++(−1)pf⊗(d′∘g)−(−1)p+qf⊗(g∘d),
whence
κ(D(f⊗g))=d′′∘f∘g−(−1)pf∘d′∘g++(−1)pf∘d′∘g−(−1)p+qf∘g∘d=d′′∘f∘g−(−1)p+qf∘g∘d=D(f∘g)=D(κ(f⊗g)).
One proves analogously that the homomorphisms (15) and (16) are morphisms of complexes.
One deduces from the morphism (14) homomorphisms of k-modules (X, p. 62)
(17) Hp(HomgrA(C′,C′′))⊗kHq(HomgrA(C,C′))→Hp+q(HomgrA(C,C′′)).
Taking C=A, one sees that the homomorphism :
(18) HomgrA(C′,C′′)⊗kC′→C′′ which maps f⊗x to f(x) is a morphism of complexes of left A-modules ; there is associated to it a canonical homomorphism (X, p. 80) of graded A-modules γ:H(HomgrA(C′,C′′))⊗kH(C′)→H(C′′), which corresponds to the canonical homomorphism of k-modules
λ:H(HomgrA(C′,C′′))→HomgrA(H(C′),H(C′′))
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.
Nothing.
Nothing.