Foxshield

Gary gives the daily and monthly costs of retaining these hirelings, as well as short explanations of what they do.

Gary also includes a great mechanic for commissioning something from a hireling: for a retained hireling to create something specific, it costs, on top of the cost of retaining the hireling, 10% of the standard cost of an item. This is a neat way of working through the problem of, 'there are no longswords for sale in this town,' a way that certainly costs money, but cuts down on paperwork and hassle. I don't remember this (or an analog) being in ACKS, but it's such a short mechanic to write down that it's quite possible that I just missed it. The only problem with this that I see is that it's important to know how long it will take to make whatever is being commissioned.

Beginning this section, Gary spells out the difference between henchmen and hirelings: henchmen are the PCs' followers while hirelings are the PCs' employees. This is a great, succinct way of phrasing the difference that I'll probably use.

Gary lists fifteen types of expert hireling, with one entry– mercenary soldier– distilled into 19 types, along with their monthly upkeep cost; only non-officer soldiers are available for daily hire, and they charge a month's pay for hazardous work, no matter the length of time.

Gary goes on to describe these hirelings. One thing of particular note is that certain hirelings will be outright necessary to retain once a PC goes about establishing a stronghold. Obviously, men-at-arms will be needed, though for most classes a certain number of these show up at the stronghold without effort on the PC's part (but do need to be paid). In addition, however, each group of 40 soldiers needs an armorer to maintain them, each group of 40/160 men need a blacksmith to maintain them (only the first blacksmith is limited to maintaining 40 men for some reason) and every 80 men need a weapon maker to maintain them (and Gary suggests dividing weapon makers into those dealing with archery, those dealing with swords and daggers and those dealing with everything else, potentially doubling or tripling the number of weapon makers needed to maintain 80 men). These hirelings would work full-time on maintaining their men, not having any time for commissions from their PC employer; for commissions, the PC will need to retain a hireling caring for less than the full number of men they can maintain. In addition, the wise PC will retain an engineer-architect for any serious building, or risk the building falling over in 1d% months. A PC will likely also want to employ a steward/castellan, who will not actually do the work of establishing a stronghold for the PC but will maintain, run and defend the stronghold, even in the absence of the PC.
One thing I should point out that I liked is that Gary has the DM roll for the skill level of a few of the hirelings, like the jeweler, armorer and sage. Skill levels for armorers and sages are known to players, but skill levels of jewelers (and thus the chances of wonderful success and utter failure) are not.
Finally, Gary spends by far the most space in this section on sages, the walking, talking, 'encyclopedias, computers, expert opinions and sort of demi-oracles of the milieu all rolled into one.' Sages have a major field of study, of which they specialize in two or more special categories, and one or two minor fields of study. For the most part, PCs will retain sages to answer questions. Gary provides charts to determine the chance that the sage will be able to find the answer to the question based on how general or specific it is and whether it is in one of the sage's areas of special study or not, as well as the cost of the research and the costs of maintaining the sage, which include providing for up to 100,000 gp worth of material relating to the sage's areas of study, research grants and even more expenditure if the PC commissions the sage to master new areas of study. Sages, unlike other expert hirelings, may be consulted before a PC reaches name level and establishes a stronghold. My speculation is that Gary put so much work into sages because his players were always asking questions about things in his game; since Gary included the sage as a mechanism for them to have their questions answered, he could get them to stop asking him all their questions out of character and point them to the sages. I like these sages a lot; my players aren't so persistant as I'm guessing that Gary's were, but I like having a way to give my players information when they really want it without having to mess with what the characters in the game world should be able to know.