Old English plant names do not correspond one to one with modern species, and the medical manuscripts are not consistent with each other. Mucgwyrt, wegbrade, fille and finule can be identified with reasonable confidence. Maegde and attorlathe have proposals that have not been accepted. Stune and wergulu have no identification that anyone defends, and nobody holds a position on them, so the question is an unfilled gap rather than a live argument. Every attempt to read the charm as pharmacology stops here.
What would resolve this A glossary witness pairing one of the unidentified Old English names with a Latin plant name, which is how the secure identifications were made. Failing that, archaeobotanical evidence from a securely dated Anglo-Saxon medical or monastic context, where the plant assemblage could be compared against the list. A systematic re-search of the Old English glossaries and of the English place-name corpus for the elements stune and wergulu, using material catalogued since the standard identifications were made, is the concrete next step.
OpenOld English lexicography, historical botany and archaeobotanyOswald Cockayne, Leechdoms, Wortcunning and Starcraft of Early England