You'll probably want to go with the former, incendiarius, as that is the Classically-attested word. It's actually an adjective, but it's used as a noun (meaning "arsonist") in Tacitus and Suetonius.
On the other hand, incendarius is a Medieval variant.
It might also be of interest for you that arsonist is a relatively recent coinage; the older word ...
Five years later, I return to give a different answer.
In my previous answer, I claimed:
In Classical Latin, there were no words exactly corresponding to "yes" and "no". Non and ne were negatives, but they needed to combine with other words (like "not" in English).
This is what I learned for formal Ciceronian or Caesarian ...
The use of homo as a genuine indefinite pronoun is definitely not good Classical Latin, but homo did have indefinite meaning from the very beginning, and by the Christian era Vulgar Latin was clearly well on its way to developing it into a true pronoun (cf. the Vulgate's non in solo pane vivit homo, often translated as "one does not live on bread alone&...
We would need more context (especially the text before your quotation) to be sure. But my preliminary reading would indeed be that she felt it was not "up to her" to do it. Needless to say, the expression would not be written this way in classical Latin.
I can't find any evidence of Latin using homo as an impersonal pronoun. It might be interesting to note that the nominative case of Old French hom becomes on, which is used as an impersonal pronoun in French, while the oblique home became homme, meaning man.
As for Latin, the passive voice was used instead of any impersonal pronoun. This meant that for ...
The Latin for "constellation" is (surprise surprise) "constellatio". It is not exactly classical, though L/S do have a citation from Ammianus Marcellinus (4th century). In older sources we do find "astrum", but this can also be a single star, and not necessarily a constellation. "Constellatio" is unambiguous.
Original attempt: sub lux astri lucens
Your choice of vocabulary is good, and the word order is natural.
What you are missing is some inflection.
There is no freedom of choice when it comes to case, number, and gender here.
Rather than giving you the full answer, I will give you a list of specific questions, as I believe this to be most useful for learning.
Thesaurus linguae Latinae (TLL) is a Latin to Latin dictionary by the Bavarian Academy of Sciences and Humanities certainly has a prominent place in the list, in that it is the most exhaustive dictionary of Latin, and it is available freely online.
The -really relative- handicaps are:
as of July 2021 it's currently only up to the letter R,
it is projected ...
Aliquot as an indefinite adjective is common and is very suitable for your purpose.
Here are some examples taken from archaic and classical authors:
Cato, orig. fr. 128 Peter: interim aliquot <p>au<ca> castra feci
Plaut. capt. 161: eorum sunt aliquot genera Pistorensium
Cic. Cluent. 168: dico illum [...] aliquot dies aegrotasse et ita
Searching "aliquot castra" I happen by luck to find an old article (Rand, Edward Kennard. "On a Passage in Virgil's First Eclogue." The Classical Journal 2.3 (1907): 125-128.). in which the issue at hand is mentioned (tersely unfortunately).
In that paper a passage from the Ecoluge is dealt. For our need however, the relevant line is:
The Latin–Norwegian Latinsk ordbok : latin–norsk is available both in print an as an e-edition for those registered at a Norwegian university. But previous editions are available for free via Nasjonalbiblioteket for Norwegian IP adresses. It is a standard work used at Norwegian universities and sixth-forms.
It has a very long history, starting as ‘...