v1.0.0 — Author-direct dataset (194 seals, 264 signs)

First citable release.

Datasets (traceable to author source documents):

• data/seed/seals_master.csv — 194 seal readings in Tamil with romanization
• data/seed/signs_master.csv — 264 sign-to-phoneme mappings (5 rows flagged pending author confirmation: 39, 51, 52, 322, 391)

Layer 1 (phonemes) complete; Layer 2 (English meanings): 20 complete, remainder in progress.

External Parpola/CISI corpus removed — all data is author-direct. See docs/VALIDATION_REBUILD_REPORT.md.

Cite as: Shanmugham, P. (2026). Open Indus Lab, v1.0.0.