It's a failure of the game mechanics to maintain consistency with the setting as written.
As a house rule, it's fine, but it's a patch over a threadbare portion of jeans. Patches are uncomfortable and itchy. I'm going to go off on a tangent here.
Ritual magic is kind of the odd duck of Ars Magica. It takes more time than formulaic spells, but that time can easily be accounted for within the backdrop of a season. Rituals are rarely cast under some kind of time constraint, because the game doesn't really encourage time constraints (the example of penetration in the core book is exemplary of this approach), so the amount of time a ritual takes is mostly fluff. I think rituals need to cost something, and the most obvious thing is that they cost time, like a season. Such a mechanic in Ars isn't unprecedented, HoH:TL introduced Fenicil's rituals, which have Greater and Lesser versions. Greater rituals require an entire season to enact. Healing rituals would still be lesser rituals. Rituals which provide some ongoing benefit (I'm looking at you stat boosting rituals) or create something out of whole cloth need an entire season to enact.
Granted, casting an Aegis as a seasonal activity each year is a tall order. So maybe it lasts longer than a year. Maybe it is 7 years (Notatus invented it, after all). One season out of 28 isn't a tall order, in one covenant my character has to give up 1 season out of 6 to covenant service, if he could do this as one of his seasons of covenant service, he'd jump at it. Alternatively, rituals like the Aegis, and Shrouded Glen might last for a number of years equal to the Magic Theory score... Young magi in a spring covenant can only power their Aegis for a few years, but the Bonisagus of Durenmar have an Aegis that lasts 14 years or more. A lot of magi would be able to power the Aegis for 6 years...
Now that we've determined the time it takes (and how long the ritual endures in the case of the Aegis). What is the casting total then? I'd lean towards the lab total, as it is almost always going to be higher (because int is usually higher than stamina, and magic theory, a tricked out lab, etc), but also include AL and Philosophiae. I'd even leave the aura in the final calculation for penetration purposes, just to account for the "home field" advantage magi have in this particular aura on the grounds covered by the Aegis. Penetration would still be lab total- effect level, so ideally you're still looking to double the effect level as if inventing a spell or instilling an effect in one season. Vis cost would be based on the magnitude of the final effect level. Since it's not just a lab activity, I'd make the limit on vis used be a combination of the techniques and forms, each limited by the magic theory score. So a 30th level Aegis costs 6 pawns of vis, and could be accomplished by someone with a Magic Theory score of 3, at the cost of 3 pawns of Vim AND 3 pawns of Rego. If the magic theory score were 4, he could use 4 of one and 2 of the other. If he needed vis to boost the ritual's penetration, he could use 4 of each technique and form and get a bonus to the final penetration.
Just spit balling here...