The bulk of this entry comes from commentary to the two Aristophanes passages quoted below. Both within this entry and in the preserved attestations of the headword in general there is much alternation between the variant spellings of the headword noun, skinda/lamos and skindalmo/s. Luckily this entry is free from another attested variation between skinda/lamos- skindalmo/s and sxinda/lamos- sxindalmo/s. (Moeris sigma50 identifies the latter as an Atticism.)
[1] = Synagoge sigma123 (with the variant spelling skindalmoi/); Photius, Lexicon sigma333 Theodoridis; for the second gloss cf. ps.-Herodian (On Orthography 3.2.581), Hesychius sigma1008.
[2] A curious statement, since the most recently mentioned form of the word, skinda/lamoi, is nominative and proparoxytone. In general, not just the nominative but all forms of the variant skindalmo/s are oxytone (apart from genitive and dative plural which are perispomenon), whereas only the oblique forms of the variant skinda/lamoi (apart from accusative singular) are paroxytone. It is as if the commentator thinks that certain forms of the two variants form a single, hybrid declension.
[3] cf. scholia to Lucian (67.5 Rabe).
[4] Aristophanes, Clouds 130.
[5] Aristophanes, Frogs 819, exhibiting essentially the readings of the mss and the scholia (with metrically inconsequential variations between sxindala/mwn and skindalmw=n, etc.). In his OCT Wilson adopts a conjecture of Dover reversing the case of the nouns: skinda/lamoi te paraconi/wn: "splinters of linchpins".