Thanks, Eric. That accounts for the raptorial forelegs, which seemed out of character for a phasmid. But assassin bug never occurred to me. Jerri Larsson from Bioquip suggests that this may be E. brevicoxa, which is the most common of the thread-legged bugs in California.

With all due respect to Jerry, I found the following:

In 1909, Banks described two species of genus Emesa. Genus Emesa was later renamed to become Emeseya; see below.
Banks stated that Emesa longipes (Say) is common in the Eastern U.S. [Note: Emesa longipes = Emesaya brevipennis.]
He introduced Emesa brevicoxa as a new species in 1909 from a short description of a single female from 'Los Angeles'.
He describes the key difference between the Emesa brevicoxa female and specimens of Emesaya brevipennis as the length of their fore coxa:
* Emesa longipes: fore coxa "over twice as long as the head", and
* Emesa brevicoxa: fore coxa is "hardly twice as long as the head."
Note that this distinction yields little to no decisive separation space.

One year later, in his short publication on the Emesidae (now Emesinae) of California, Baker listed only one Emesa species as occurring in California:
Emesa brevicoxa: "Our largest and most abundant [Emesidae] species. Often congregates in large numbers about barns and sheds, groups frequently resting in one place with very little movement, for many days consecutively, in the Fall."

McAtee and Malloch changed the generic name from Emesa to Emesaya in their extensive 1925 revision of American Reduviid subfamily Ploiariinae.
They introduced three subspecies of Emesaya brevipennis:
1. Emesaya brevipennis brevipennis distributed from the Eastern U.S. to Texas,
2. Emesaya brevipennis australis distributed from the U.S. Gulf States south to Panama, and
3. Emesaya brevipennis occidentalis distributed from (Alta) California and Baja California.
They also wrote that "Emesaya brevicoxa is only described from a single female and its taxonomic status is unclear."
This was also the conclusion of the extensive (624 pp.) work of Wygodzinsky on Emesinae in 1966.
Wygodzinsky also validated the three subspecies as described by McAtee and Malloch.

It seems that, as Emesaya brevicoxa was never described from more than one specimen, it has likely been 'covered' by Emesaya brevipennis occidentalis.

B.t.w., the specimen you found looks like a male, although it's hard to tell from the photos.
In any case, its fore coxa are marginally longer than twice the length of its head, which, if it were a female, would still render it Emesaya brevipennis occidentalis.

in a 2001 Florida Entomologist PDF doc here, the authors cite Usinger's 1941 paper ("Rediscovery of Emesaya brevicoxa and its occurrence in the webs of spiders") which apparently states that several specimens of brevicoxa were found in cobwebs beneath the eaves of a cabin. The three authors of the 2001 paper (one of whom is J.D. Bradshaw at BugGuide) apparently accept the distinction that bevipennis will eat insects in captivity whereas brevicoxa will eat only spiders in captivity.

Whatever the status of brevicoxa turns out to be, until someone comes up with a reliable way to distinguish it, I think the safe thing to do would be to move this photo up to the genus level.

It was hard to find a copy of the paper, of which the text is less than two pages long.
The only other 'new' difference Robert Usinger mentions between E. brevicoxa and E. brevipennis, besides their dietary preference as noted above, is that they were observed moving up-and-down while hanging in spider webs, whereas E. brevipennis has been reported to swing back-and-forth ...

So yes, moving it up to the genus level seems appropriate until more is learned about their potential differences.