I have mixed feelings about this.
On one hand settling uMgungdolovu was wisely given priority.
On the other, tech choices...
Masonry without attempting either wonder(for speeding walls? - could at least get it early and gift stone to HC to get the mids/GW built closer), and both HBR and IW?
Thus pottery was delayed, along with writing.
Given all that the economy is in better shape than it just might.
Seems to me CoL needs be the first tech priority. Ikhandas are nice enough might let one get away with an extra city early (or tech to CoL faster) but are no substitute for courthouses if the aim is to expand beyond 6-7 cities.
You certainly shouldnt start a war without it.
And if constuction isnt far off its worth waiting for.
Also there's the matter of proper timing to be brought up. Snatching the first 2 workers rather than building them can prove game-breaking, thus justifing the risk of reprisals and likely doomed diplo potential. The same thing in 500BC with a handfull workers of your own fielded is plain silly.
I swear first time through i was confused taking "this very tempting situation"
to refer to a potential city site to settle.
Same thing about combo impi+HA skirmishes. There is a proper time for it and around 150BC plus to build some HA its now past.
BTW, was the composition of the current military panned? Archers are often needed to defend early barbs, but 7 promoted for field duties rathen than defence?
Last but not least are we sure HC's land is the proper place to expand to?
I like leaving an industrius or tech-confident leader as my later target(still pre rifles/grens era though), excpecting to get more from him.
Not to mention there appears to be great land to uMgungdolovu's NE including 3 gems and several calendar commerce resources to be had. I'd aim the barb city there and claim that land over HC's, even if it likely gets claimed by someone else. Distance should be less of a consideration with only 30% maintenance eventually.
PS: Trade for both maths and alphabet ofcourse, just try to do it with AI's low in tech/tech potential and likely to be your first victim.