It would be worth distinguishing between the Sira (i.e. biographical texts of the prophet Muhammad like ibn Hisham/Ibn Ishaq and Tabari) and collections of hadith (i.e. collected sayings and actions of the prophet as reported through a chain of transmitters, like the six "sound books" of Bukhari, Muslim, etc.) The former was in part composed of the latter but without the chains of transmissions and to a lower standard of rigor even by medieval standards and also included the maghazi (tales of exploits in war) to such an extent that the entire genre is sometimes termed Sira/Maghazi.

The validity of using these sources is more or less the problem in early Islamic historiography. The problems are basically two fold. We already know they weren't written down until a couple centuries after the fact, so even if everyone was being truthful in their chains of transmission then we're dealing with a 200 year-long game of 'Telephone'. But, to make things worse, we know that there was a huge amount of forging of these hadith going on at the time. That was more or less the point of collecting the 'sound' hadith into these books and in writing the enormous biographical dictionaries that pop up around the same time with judgements about the individual transmitters.

There are different approaches to this. The traditional western historical method is sometimes termed the "kernel of truth" approach, which posits that while the hadith and the sira might have accrued a lot of extraneous material the basic facts therein can be regarded as accurate. W. Montgomery Watt's biographies of Muhammad are probably the example ne plus ultra of this approach. Work by Patricia Crone and others from the 1970s onwards, however, have seriously challenged the notion that such a thing is possible and instead claim that the hadith and sira must be rejected entirely as unreliable.

A new problem arises in trying to compose a history of early Islam or of the life the prophet without these sources, which are termed the "traditional sources" (that is, they are based on oral traditions), which is that we have very few alternative contemporary documentary sources that could replace them. Among the Islamic sources these are basically limited to the Quran, the Constitution of Medina, a handful of very early inscriptions and later sources from after the conquests, particularly things like papyri and coinage. All of these are of very limited value in actually telling us about the life of Muhammad or early Islam, or where they are valuable it seems to contradict much of what we know. Another approach is to use non-Muslim documentary sources, which likewise produces some very odd and controversial results for those who are more used to the traditional accounts (for example, see Crone's Hagarism.)

The best overview of this historiography that I'm familiar with is Robert Hoyland's Writing the Biography of the Prophet Muhammad: Problems and Solutions.