The guidelines of the Ancient Greek Dependency Treebank 2.0 have been written to annotate Ancient Greek texts.The epigraphic texts, however, pose a challenge for those carrying out morphosyntactic annotation: should we remain as close as possible to the actual epigraphic text, or represent it in an interpreted and normalized version?How should all epigraphic peculiarities which do not have standard editorial representation, such as, for example, punctuation marks, be treated?A small corpus such as that of the inscriptions of the Euboean colonies of Sicily of the archaic and classical period has allowed us to test different options and evaluate the annotation challenges.This contribution is the result of a discussion about the advantages and disadvantages of often opposed annotation possibilities.We present here our first proposal for an adaptation of the guidelines to analyse the morphosyntax of inscriptions, which we hope will stimulate discussion between epigraphists and linguists.In particular, we propose to try to stick to the epigraphic evidence as far as possible and therefore render its complexities (e.g., local alphabets, dialect variants not attested in literary texts, ellipsis, punctuation marks, and word forms which can be linguistically interpreted differently), while trying to preserve consistency with the annotation of literary texts.