TL;DR: Different authors have different conventions. E.g. the covariant derivative is $D_{\mu}=d_{\mu}\pm iqA_{\mu}$. Nevertheless, everybody seems to agree that the Dirac Lagrangian density is ${\cal L}=i\bar{\psi}\gamma^{\mu}D_{\mu}\psi+\ldots$, so the interaction term becomes ${\cal L}_{\rm int}=\mp q\bar{\psi}\gamma^{\mu}A_{\mu}\psi$, and hence the Feynman vertex rule is $\mp iq\gamma^{\mu}$, respectively, i.e. it comes down to a sign ambiguity.
For a correlation function with an odd number of external photon legs, this yields an absolute sign ambiguity (but no relative sign ambiguities), and hence no physical consequences for probabilities if we flip the overall sign of all the charges $q\to -q$.
Concerning the references listed by OP it goes as follows:
D. Griffiths, Introduction to elementary particles, has
Minkowski sign convention $(+,-,-,-)$ (p. 85);
charge sign convention $e=|e|$ (p.395);
covariant derivative $D_{\mu}=d_{\mu}+iqA_{\mu}$ (p.349,352,365);
interaction term $-q\bar{\psi}\gamma^{\mu}A_{\mu}\psi$ (p.348);
and Feynman vertex rule $-iq\gamma^{\mu}$ (p.230).
M.E. Peskin and D.V Schroeder, An Introduction to QFT, has
Minkowski sign convention $(+,-,-,-)$ (p. xix);
charge sign convention $e=-|e|$ (p.809);
covariant derivative $D_{\mu}=d_{\mu}+iqA_{\mu}$ (p.303);
interaction term $-q\bar{\psi}\gamma^{\mu}A_{\mu}\psi$ (p.303);
and Feynman vertex rule $-iq\gamma^{\mu}$ (p.303,802).
M.D. Schwartz, QFT & the standard model, has
Minkowski sign convention $(+,-,-,-)$ (p.12,817);
charge sign convention $e=|e|$ (p.818);
covariant derivative $D_{\mu}=d_{\mu}-iqA_{\mu}$ (p.140,224,818);
interaction term $q\bar{\psi}\gamma^{\mu}A_{\mu}\psi$ (p.225);
and Feynman vertex rule $iq\gamma^{\mu}$ (p.226).
For more on conventions in QED, see e.g. this Phys.SE post.
--
$^1$ We work in units with $c=1=\hbar$.