Hello, somebody asked a very similar question recently to which I linked them to my thesis on the subject here. The long and short of it is that, as much as we wish there was, there wasn't one unified blueprint for how to build a burh, although there certainly were unified elements. It's likely these were decided upon at a meeting attended by Alfred, Edward, Æthelflæd and Æthelred in the 890s in London. There aren't that many sites based on Roman defences - London, Chester and Exeter are main examples - but even where they were, there was typically a major restructuring of the road network (this happens in London in the 890s according to work by Alan Vince) and a significant change of the urban layout. According to the Anglo-Saxon Chronicle and the Mercian Register, the majority of the labour at burh sites was likely carried out by the fyrd, or levied as manpower taxation from the local population as commonly stipulated in charters of land grants.

One of the most common phenomena, seen in burhs from London to Wallingford to Tamworth, is the division of the area inside the fortifications into rough quarters. Typically one of these was a cluster of productive sites, and one quarter was left largely empty. Under the old model, it was assumed that this was for cattle or refugees, but under the new model established by Baker, Brookes, Reynolds et al, it's far more likely that this area was set aside to house the rotating garrisons of fyrd soldiers, who would only be present seasonally, and would most likely be billeted in tents. In many cases, a church could be found at the centre of the settlement. In de novo centres like Stafford this could also have been originally a temporary structure.

In burhs where there wasn't pre-existing Roman fortifications, defences typically took the form of a series of ditches and palisades, connected by fortified gatehouses. Even in contexts where these defences already existed, Tamworth or Worcester for example, they were typically greatly expanded, enhanced and otherwise improved. Depending on the burh and its intended long-term purpose, these defences could take different forms. At Tamworth, Gould's work identified a series of three concentric berms, separated by ditches and topped with palisades, mutually supportive and connected by gatehouses. At Worcester, the defences were more of a 'spiral' in shape. At London and Bath, the defences included fortified bridges to deny the use of the rivers Thames and Avon Avon respectively to hostile fleets. In the case of London at the very least, the bridge was defended on the South bank of the Thames by an outer working of the fortress, or Southwark. The burh at Eddisbury was previously an extensive Iron Age hill fort, rebuilt impressively in 913 by Æthelflæd's Mercian army. The burh had no previous settlement or intended urban function, and so was divided into two simple sections, at least one of which had a bread oven cut into the ramparts. The natural steep scarp of the hill above the Cheshire Plain (the hill is quite distinctive from other vantage points across the Cheshire Plain such as Burwardsley) was further enhanced by bivalate ramparts and ditches topped with a palisade. Entrance to the fort, along a switchback incline, was controlled by a fortified gatehouse. The burh at Stafford has been less forthcoming with its defensive plan, but a series of productive sites have been excavated, including major sites of pottery, butchery and baking.