The rock shelter is located within a concavity along the foot of a small limestone cliff. The resulting vertical overhang forms an irregular elongate shelter of c. 8 m wide and c. 2.5 m deep, creating a ceiling that rises from 0.8 m in the south-east to over 2.8 m in the north-west. The surrounding limestone is a heavily eroded, poorly sorted conglomerate. Along the foot of the scarp, there is a natural terrace of gently sloping ground, forming a relatively flat apron in front of the shelter. A 6 m by 2 m trench was excavated within the rock shelter along with a second trench, measuring 6 m by 0.5 m, extending out from the rock shelter to the terrace edge. A total of 25 deposits were recorded, including hearth features and large pit. The earliest stratified cultural deposits relate to a large pit feature contained a few small animal bone fragments and pottery shards of a Bronze Age date. Pottery fragments associated with three ash deposits and a hearth were of late- or post-medieval date and overlain by a rough stone floor containing a grave cut dating to the Second World War. From the basal units of the second trench along the terrace edge over 120 struck flints were recorded. Of these, 16 were blades with an average size of 2 cm in length while medium-sized flakes, up to 2 cm in length formed the largest category, with 36 examples. Some of these pieces had been retouched to form scrapers and awls and there were also a number of burins. A number of the smallest blades (microliths) were also retouched and probably date to the Upper Mesolithic. Two core fragments, a number of retouching flakes and debitage were also recorded, suggesting that flint working may have taken place on the edge of the terrace nearby. The assemblage is indicative of active flint knapping, with the objective of the reduction strategy to produce small flakes, used as blanks for the manufacture of retouched tools. The location of the site forms a natural vantage point, sitting on the edge of the elevated terrace overlooking the valley bottom, and may well have formed a strategic location for the exploitation of wetland resources.

The 2012 excavation at Kanali rock shelter was conditioned by the results of the excavations carried out in 2004. It was decided to test two different points, one positioned near the area where the 2004 excavations were concentrated and the second point located where the collapsed entrance of the cave was thought to be. The first excavation site was named Kanali 1, while the second site was named Kanali 2. Kanali 1 In the end of this year's excavations in this point, was increased the amount of finds and also the lithic collection. From the material obtained from the excavation, it appears that the main goal of the occupation was to achieve the production of tools of the blade type and small flakes, a part of which can be described in some cases as flakes (<1.5 – 2 cm). This is evidenced by the presence of 5 cores for the production of blades as well as the blades themselves, as well as 2 discoidal cores and the corresponding flakes and blades. Other tools present in the collection are: a chip thinned in the ventral part, a retouched blade, an end scraper, and two partially preserved but unidentifiable tools. Of a different operational chain, are two bidirectional and two-sided wedges, which may have been in their initial phase. Kanali 2 On the other hand, the lithic material found during excavations at this point represents a special case in relation to the history of the study of this territory. Among the most important finds were a core, for the production of blades, as well as the presence of numerous blades. There are also two end scrapers which seem to have been produced from the same operational concept. It should be mentioned that the end scrapers are not short or round but rather long. Based on the material found from the excavation of this year, some preliminary conclusions can be given regarding the chronology of this collection. Starting from the high presence of blades with worked backs as well as the presence of a trapezoid, we can say with certainty that we are dealing with a collection of the Epigravetian, the period after the Last Maximum Glaciation (20,000 years). On the other hand, the presence of longitudinal end scrapers (not circular) as well as the absence of knives with worked backs places this material in time to the early Epigravetian or about 20,000-15,000 years ago.