Researchers present an open, fully offline digital reader for the Prasthanatrayi corpus, including Sankara's commentaries. The tool resolves continuous euphonic combination and dense compounds by making every word clickable to reveal its split, morphological analysis, and gloss.
- The resource covers 2,971 verses, sutras, and prose sections with 36,881 analysed word-occurrences of the root text.
- It includes a global dictionary of 95,587 distinct commentarial surface forms and functions as a concordance for search.
- The hybrid pipeline uses a rule-based sandhi splitter over an inflected-form lexicon with LLM-assisted analysis under an adversarial two-pass verification protocol.
- Intrinsic evaluation shows high-confidence analyses agree with an authoritative inflectional lexicon on over 99% of attested forms.
The reader is provided as a single self-contained HTML file requiring no server or network, serving as a freely redistributable teaching and reading aid.