Hoping to get more to work with for the ID here...I searched through my own 2013 BugGuide Gathering photos from this same oak tree, and found 4 images I'd gotten of a male Pepsis. That individual keyed in Hurd(1)[1952] to P. angustimarginata:

 

P. angustimarginata was the name used in Hurd(1)[1952] and other references focused on the nearctic fauna from past decades. However that name was synonymized under the name P. basifusca in the ambitious three-part revision of Pepsis by Vardy(2)[2005], where the emphasis is on the vast majority of species ranging from Mexico into South America.

After working through the ID of the male above, I had anticipated that J & J's post here might be a female of P. angustimarginata as well. I was encouraged by the narrow black border on the fore- and hind wings (angusti- = narrow; marginata = margin), and the pronounced mesopleural tubercle here. The absence of a "hair brush" on the 4th abdominal sternite here was not a problem, as females lack that distinctive character present only in males of certain Pepsis species.

But in carefully checking Hurd's key and descriptions, I found that a critical diagnostic character given there for female angustimarginata seems to be absent in the image here...namely, the "presence of erect hairs on the upper surfaces of the middle, and more particularly the hind, femora". So I carefully went through the key in Hurd(1)[1952] again, using the photo of the female posted here, as well as seven other photos J & J sent me through email (which, likewise, don't show discernible hairs on the upper surfaces of the middle or hind femora). Here's the key couplet series, as best I could determine things:

1b) Females (antennae convolute; abdomen with six visible segments).....19
19b) Anterior femora below glabrous, or at most with a few short hairs....22
22b) Spurs of middle tibiae not acutely curved near apex, straight or only slightly curved.........24
24a) Antennae with first flagellar segment equal to, or shorter than, the distance between the inner eye margins at the level of the posterior ocelli.........25
25a) Wings predominantly xanthochromatic.........26
26a) Forewings with apices wholly dark.........27
27a) Not marginata (a species with range south Florida and Caribbean).........28
28a) Antennae wholly dark.........29
28b) Antennae with at least the apical segment tipped with orange, frequently several or all flagellar segments orange.........mildei
29a) Posterior femora above glabrous or at most with a few scattered hairs; forewings with exterior marginal dark band broad, occupying at least one-half of the distance from the wing apex to the middle of the third transverse cubital vein.........30
29b) Posterior femora above armed with erect and recumbent bristle-like hairs, most noticeably distally; wings with exterior marginal dark band narrow, occupying considerably less than one-half of the distance from the wing apex to the middle of the third transverse cubital vein.........angustimarginata
30a) Wings bright fiery red..........chrysothemis chrysothemis
30b) Wings brownish yellow or fulvous red.........chrysothemis lucasii
The keying sequence above seems straight-forward up to couplet 28. There, one might conceivably argue for going with 28b) and mildei since:

1) two of the photos emailed to me indicate a tiny amount of pale orangish-brown color at the tips of the antennae; and
2) Hurd(1)[1952] and Hurd[1948] mention that the amount of orange in antenna can be quite variable, and even lacking in mildei...particularly in some AZ specimens. Below is a quote (bold emphasize mine) from the 1948 paper:
"In some specimens, both male and female, the orange or reddish orange color is limited to the apical segments or segment, being replaced in part or in full, on the basal segments, with black. There seems to be a continuous variation in this respect, particularly with specimens from the interior. [...] In some Arizona material the antennae are entirely black, but the writer does not, at this time, consider this fact significant enough to warrant varietal status."
On the other hand, Vardy(2)[2005] in his description of P. basifusca (=angustimarginata) states: "antenna sometimes becoming faintly brown apicad". Based on the photos available here, that could be taken as sufficient reason for rejecting the choice of Hurd's lead 28b) (i.e. mildei)...and instead going with 28a) above.

That takes us to couplet 29), which consists firstly of a "hairs on posterior femora" character; and secondly, of a "dark exterior marginal band of the forewing" character. The second character clearly points to 29b) and angustimarginata here...as the dark marginal band in the photos is much narrower than half the distance from the wing apex to the middle of the third transverse cubital vein (i.e. the closed cells).

But the first character of 29b) appears not to be met here...no hairs are discernible above the middle or hind femora. If we gave priority to that discrepancy we would end up at chrysothemis. But chrysothemis is a member of the "Pepsis rubra group", which Vardy(2)[2000] emphasizes is characterized for females by the character: "Radius meets costa at a very shallow angle", as illustrated in figures 76-78 on pg. 44 of Vardy(2)[2000]. And close examination of the angle at which the radius meets the costa in the photos J & J sent me indicate this is not a member of the "rubra-group"...and thus not chrysothemis.

All the above seems to indicate angustimarginata (=basifusca) as the best candidate, but again one that is not entirely satisfactory...since the hairs described by Hurd on the upper side of the mid- and hind-femora of that species seem entirely absent here. Which leaves the ID seeming somewhat unsettled.

Note: Six abdominal terga & thin, curled antennae = ♀

Moved from Spider Wasps.

Moved from Ants, Bees, Wasps and Sawflies.