The Karoo Basin covers much of South Africa and is an area of prospective shale gas exploration, with the Whitehill Formation the target shale unit. However, the sedimentary succession, including the Whitehill, has been intruded by a series of sills and dykes associated with the Karoo Large Igneous Province (~183 Ma), which are expected to have modified the thermal history of the basin dramatically. Here, we investigate a secondary effect of these intrusions: a series of hydrothermal vent complexes, or breccia pipes, focusing on using O, H, and C isotopes to constrain the origin and evolution of fluids produced during the intrusion of basaltic sills. A cluster of breccia pipes have been eroded down to the level of the Ecca Group at Luiperdskop on the western edge of the Karoo basin; a small isolated pipe of similar appearance crops out 13 km to the east. The Luiperdskop pipes are underlain by a Karoo dolerite sill that is assumed to provide the heat driving fluidization. The pipes consist of fine‐grained matrix and about 8% clasts, on average, of mostly sedimentary material; occasional large rafts of quartzite and dolerite are also present. The presence of clasts apparently from the Dwyka Group is consistent with the depth of formation of the pipes being at, or near, the base of the Karoo Supergroup, between 400 and 850 m below present surface. The presence of chlorite as the dominant hydrous mineral is consistent with an emplacement temperature between 300 and 350°C. The major and trace element, and O‐ and H‐isotope composition of the Tankwa breccias is homogenous, consistent with them being derived from the same source. The δ18O values (vsVSMOW) of the breccias are relatively uniform (7.1‰–8.7‰), and are similar to that of the country rock shale, and both are lower than expected for shale. The water content of the breccia is between 2.7 and 3.1 wt.% and the δD values range from −109‰ to −144‰. Calcite in vesicles has δ13C and δ18O (VSMOW) values of −4.2‰ and 24.0‰, respectively. The low δD value of the breccia rocks does not appear to be due to the presence of methane in the fluid. Instead, it is proposed that low δD and δ18O values are the result of the fluid being derived from the breakdown of clay minerals that formed and were deposited at a time of cold climate at ~290 Ma.