# Kosha public dataset catalog

> Public-tier datasets of the kosha dataset manifest — title, license, size and format per dataset, for machine consumption.

Generated from the kosha dataset manifest `data/manifest/datasets.json` (https://github.com/gasyoun/kosha); public tier only — intermediate and restricted datasets are deliberately excluded. Regenerate with `python tools/gen_llms_txt.py --catalog <kosha-checkout>/data/manifest/datasets.json` `--out data/llms_catalog.txt`; verify byte-stability with `--check`.

## Datasets

- [Amarakosha varga/synsets x MW/PWG sense-unit inventories (structural split envelope)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 5590 rows; format=tsv
- [ashtadhyayi.com site dataset — Aṣṭādhyāyī (sutraani/) + Dhātupāṭha (dhatu/)](https://github.com/ashtadhyayi-com/data): unspecified; 2259 rows; format=json
- [BHS (Edgerton) × MW/PWG × DCS Buddhist loci key2 crosswalk](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 18188 rows; format=tsv
- [Bloomfield 1906 Vedic Concordance — direct Ṛgveda citations (pratīka index)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 36680 rows; format=tsv
- [Bloomfield 1906 RV citations × Elizarenkova Russian RV — citation-to-translation pointer join](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 36680 rows; format=tsv
- [Pan-CDSL headword character-n-gram oracle (allngramtxt.txt)](https://github.com/sanskrit-lexicon/CORRECTIONS): CC BY-SA 4.0; 6656616 rows; format=txt
- [Start chteniya cohort pack freeze (Hitopadesa-0 + subhashita-beginner + sandhi L1-3)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 5 rows; format=json+tsv
- [Corpus sandhi — frequency-ranked rules across 41 texts](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 13012 rows; format=tsv
- [Correction-event log, full trio (all/typed/final)](https://github.com/sanskrit-lexicon/csl-observatory): CC BY 4.0; 52498 rows; format=csv; doi:10.5281/zenodo.21346705
- [Correction loci — locus-resolved correction records across all csl-corrections change files (v1.0)](https://github.com/sanskrit-lexicon/csl-corrections): CC BY-SA 4.0; 39540 rows; format=tsv
- [CDSL GitHub maintenance ecosystem: A15 analysis inputs (76-repo snapshot, issue/commit ledgers, 52,498-event correction ledger)](https://github.com/sanskrit-lexicon/csl-observatory): CC BY 4.0; 52498 rows; format=csv+json (67-file catalog); doi:10.5281/zenodo.22853818
- [Cyrillic proper-noun -> SLP1 validated lookup table (H3985, SL GAPS §6)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 534 rows; format=tsv
- [Dataset x discipline join — kosha registry through the IndologyScholars meso crosswalk (H4737)](https://github.com/gasyoun/IndologyScholars): CC-BY-4.0 (IndologyScholars derived exports; archive attribution); 132 rows; format=json
- [DCS lemma ↔ CDSL headword crosswalk](https://github.com/sanskrit-lexicon/csl-apidev): CC BY-SA 4.0; 98606 rows; format=tsv; doi:10.5281/zenodo.22105641
- [DCS compound (samāsa) dictionary derivation](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 613758 rows; format=csv
- [DCS akshara/varṇa/ligature frequency tables](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 8394 rows; format=csv
- [Sanskrit lexical semantic change scores (PPMI-cosine, DCS 5-slot chronology)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 3049 rows; format=tsv
- [DCS nominal pooled-class split — long-ī and -ant/-at, on an external lexical signal](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 279 rows; format=json
- [DCS nominal pooled-class split, ADJ half — the -ant surplus G2 could not see](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 1880 rows; format=json
- [DCS nominal paradigm — case × number grid per declension class](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 14 rows; format=json
- [DCS parallel-passage alignments, full-size export (PARA/Polnorazmernye)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 506787 rows; format=csv
- [DCS reading pack — Hitopadesa, Prastavika (opening)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 125 rows; format=json
- [DCS reading pack — Kiratarjuniya 1 (Bharavi)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 92 rows; format=json
- [DCS reading pack — Nala 1 (Nalopākhyāna, Mahābhārata 3.50)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 65 rows; format=json
- [DCS reading pack — Nala 2 (Nalopakhyana, Mahabharata 3.51)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 61 rows; format=json
- [DCS reading pack — Nala 3 (Nalopakhyana, Mahabharata 3.52)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 51 rows; format=json
- [DCS syntagmatic tables of frequent lexical cores, by historical period (Приложение 6)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 19083 rows; format=csv
- [DCS syntagmatic table — all corpus lemmas (Приложение 7)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 82799 rows; format=csv
- [Sanskrit-stem co-occurrence table (full)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 353351 rows; format=csv
- [DCS prefixed-verb decomposition (MG 2014 workbook) — in-form × prefix × dhātu × Palsule №](#dcs-upasarga-prefixus): CC BY-SA 4.0; 6425 rows; format=tsv
- [DCS verb-form frequency by class, with prefixed forms](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 8454 rows; format=csv
- [DCS verb-form frequency in corpus (preliminary)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 106 rows; format=csv
- [DCS verb roots grouped by class (with attested forms)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 463 rows; format=csv
- [DCS verbal paradigm — per-root attested finite + non-finite cells](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 7689 rows; format=json
- [Definition-generation eval: Heritage (Huet) French glosses as an independent second reference (H2408)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 3742 rows; format=tsv + jsonl + json
- [CDSL headword ↔ DCS corpus concordance (B1, concordance core)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 74520 rows; format=tsv
- [Bhagavadgita etymology notes](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 101 rows; format=tsv
- [Bhagavadgita gold word-by-word analysis (all 18 adhyayas)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 9092 rows; format=tsv
- [Gita gold master lemma/POS vs DCS own tagging — divergence map](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 9144 rows; format=tsv
- [Gita inflection-engine QA — divergence ledger (E1 attested-corpus check)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 661 rows; format=tsv
- [Bhagavadgita gold morphology + compound dataset](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 9091 rows; format=tsv
- [Bhagavadgita interlinear prose paraphrase (Prose sheet)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 653 rows; format=tsv
- [Bhagavadgita reading packs — all 18 adhyayas (gold)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 701 rows; format=js
- [Gita reading-pack difficulty -- reduced 3-axis ordering (W2a follow-up)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 18 rows; format=tsv
- [Bhagavadgita sandhi — frequency-ranked rules](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 161 rows; format=tsv
- [Grammar Lab g1 topics x vidyut derivation failures, corpus-frequency weighted (H4795)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 32 rows; format=tsv + md
- [Grammar Lab Wave-1 topic graph (Whitney + Zalizniak root alternation / verbal morphology)](https://github.com/gasyoun/SanskritGrammar): CC BY-SA 4.0; 32 rows; format=json + tsv + yml
- [Handoff lifecycle gold standard (170 adversarially-verified verdicts)](https://github.com/gasyoun/Uprava): CC BY-SA 4.0; 170 rows; format=jsonl
- [Headword pairwise-overlap matrix (15-dict union)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 105 rows; format=tsv
- [Headword pairwise-overlap matrix, Tamil-fold extension (18 dicts)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 153 rows; format=tsv
- [Hitopadeśa per-text word concordance pilot — every word of the text feeding the dictionary (Tamilex corpus-dictionary pattern, H4034)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 7857 rows; format=tsv
- [Böhtlingk Indische Sprüche subhāṣitas](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 7537 rows; format=jsonl
- [Kochergina 1987 (third-party RU-SA) headword ↔ DCS corpus concordance — read-only sibling of the B1 family (H4748)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 20651 rows; format=tsv
- [DCS proper-name compound splits with frequency vectors (Kompozity names.csv)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 168879 rows; format=csv
- [DCS lemma frequency sidecar (whole-corpus + per-period)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 83277 rows; format=tsv; doi:10.5281/zenodo.22105641
- [kosha combined mastery schedule (five drill families, H3742)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 35517 rows; format=json
- [DCS per-sense frequency sidecar (3-layer: WN synset / MW sense / semdom; attested WordSem + estimated MFS)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 116788 rows; format=tsv
- [SRS deck — Rung B1 demo (Nala 1 content vocabulary, frequency-ordered)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 164 rows; format=json
- [LES-1990 («Лингвистический энциклопедический словарь», Ярцева 1990) static HTML dump](https://github.com/gasyoun/les-1990): unspecified; 1408 rows; format=html
- [CDSL LIFT pilot export (fourth serialization, H721)](https://github.com/sanskrit-lexicon/csl-standards): CC BY-SA 4.0; 250 rows; format=lift (XML)
- [Two-tier ligature canon: union of four inventories + coverage curve](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 1082 rows; format=csv+tsv
- [Markup-tag frequency census (44 Cologne v02 dicts)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 671 rows; format=tsv
- [MBh Nīlakaṇṭha vulgate × BORI critical-edition presence verdicts (H2845)](https://github.com/sanskrit-lexicon/csl-atlas): CC BY-SA 4.0; 83971 rows; format=csv
- [CDSL MDF pilot export (50-case neutral-model slice, MW-focused)](https://github.com/sanskrit-lexicon/csl-standards): CC BY-SA 4.0; 250 rows; format=mdf (SFM, one record per file)
- [DCS morphological-form frequency table (difficulty-scorer signal)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 840 rows; format=tsv
- [Generated↔attested morphology audit (A3, W1b)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 401368 rows; format=tsv
- [Generated↔attested morphology audit over `inflections` — the non-circular side (A3, H3782)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 238312 rows; format=tsv
- [Graded morphology curriculum — corpus-attested paradigms, ordered teaching syllabus](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 7134 rows; format=tsv
- [Morphology drills — fill/match practice items, evidence-linked to a DCS corpus locus](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 12000 rows; format=json
- [csl-inflect give-back candidates — the defensible subset of A¬G (A3, H3782)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 5224 rows; format=tsv
- [MW definition-generation eval — frozen 500-headword sample + attestations + scored baseline outputs (H730)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 500 rows; format=tsv+jsonl
- [Headword→root derivation table (Pāṇinian, 10 dicts)](https://github.com/sanskrit-lexicon/csl-orig): CC BY-SA 4.0; 9377 rows; format=tsv; doi:10.5281/zenodo.22105641
- [MW ↔ Heritage (INRIA) entry-level crosswalk](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 185803 rows; format=tsv; doi:10.5281/zenodo.22105641
- [MW verbal-root inventory](https://github.com/sanskrit-lexicon/csl-orig): CC BY-SA 4.0; 2113 rows; format=tsv; doi:10.5281/zenodo.22105641
- [MW senses x DCS citations — full-volume join (per-sense-freq axis)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 303023 rows; format=tsv.gz
- [vidyut-prakriya derivation harness over the AG bucket (A4, W2a)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 401368 rows; format=tsv
- [Pāṇinian sūtra ↔ DCS corpus concordance (A4, W2b)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 893482 rows; format=tsv
- [Aṣṭādhyāyī sūtra-coverage / dark-class map (A4, W3a)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 3983 rows; format=tsv
- [Bloomfield-style parallel-passage concordance across the DCS corpus (B3)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 153045 rows; format=tsv
- [Poona Dictionary × DCS corpus coverage crosswalk (H1336)](https://github.com/sanskrit-lexicon/csl-atlas): CC BY-SA 4.0; 118 rows; format=tsv
- [Curated pronoun-paradigm corrections (E1/W4 fix)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 208 rows; format=tsv
- [PW × PWG Kurzfassung cut measure — sense-block diff + pwg_ru cut queue (H4805)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 200234 rows; format=tsv
- [PWG DE edition graph — OntoLex-Lemon + TEI Lex-0 sidecars](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 11581 rows; format=turtle + tei-xml + json; doi:10.5281/zenodo.21961709
- [PWG \[Cologne Addition\] pwgbib entries for the 215 true-unknown Andhrabharati ls strings](https://github.com/sanskrit-lexicon/csl-observatory): CC BY-SA 4.0; 215 rows; format=tsv
- [PWG <ls> citation counts per bibliography abbreviation (work-family rollup)](https://github.com/sanskrit-lexicon/csl-observatory): CC BY-SA 4.0; 5683 rows; format=tsv
- [MW/AP90 sense units vs the PWG-family store: matched / unalignable / absent-candidate verdicts (wave 4)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 2767 rows; format=jsonl
- [PWG entry L → printed volume/column (print-scan anchor table, H3457)](https://github.com/sanskrit-lexicon/csl-orig): CC BY-SA 4.0; 122730 rows; format=tsv
- [PWG literary-source scan-indexing campaign registry (2025-2026)](https://github.com/sanskrit-lexicon/csl-observatory): CC BY-SA 4.0; 80 rows; format=tsv
- [PWG per-sense attestation window (Ceiling C2 phase 1)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 53003 rows; format=jsonl
- [PWG German–Russian translation memory (canonical v1 four-format pack)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 2392 rows; format=jsonl + tmx + tei-xml + turtle; doi:10.5281/zenodo.21932900
- [Curated per-work dating table for PWG's cited works (Ceiling C2 phase 2)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 45 rows; format=json
- [Reading-pack difficulty scores + graded ordering (W2a)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 5 rows; format=tsv
- [Reading-pack per-verse metre annotation (W3a)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 1095 rows; format=tsv
- [Roots frequency + attestation curriculum (W2b)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 629 rows; format=tsv
- [Inline Sanskrit→Russian gloss layer (reading packs, W-RU-a)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 2958 rows; format=tsv
- [Samasa (compound) analysis trainer](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 759 rows; format=tsv
- [Graded sandhi curriculum — ordered teaching syllabus](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 2634 rows; format=tsv
- [Sandhi drills — join/split/identify practice items](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 399 rows; format=json
- [Sanskrit root x preverb (upasarga) semantics](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 214 rows; format=tsv
- [SanskritGrammar Whitney-no x DCS lemma x vidyut dhatu crosswalk](https://github.com/gasyoun/SanskritGrammar): unspecified; 876 rows; format=csv
- [Sanskrityatina 04_Reverse coding legend — 187,992-headword reverse-dictionary stock taxonomy (H4475)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 27 rows; format=tsv
- [Sanskrityatina reverse-index full stock (266,703 IAST headwords) joined to the 16-dict union headwords — 91.56% coverage, 22,508-form Russian-school-only residue (H4800)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 266703 rows; format=tsv
- [Sanskrityatina 04_Reverse/Experimental SCH research cases — six-file keyed TSV with L/pc page refs (H4533)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 16304 rows; format=tsv
- [Scherzl case-government relations × DCS-treebank adjudication](https://github.com/gasyoun/SanskritGrammar): CC BY-SA 4.0; 1168 rows; format=tsv
- [semdom <-> Amarakosha crosswalk (SIL semantic domains <-> AK vargas + synsets, ID pairs)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 5898 rows; format=csv (Level A varga map) + tsv (Level B candidates, gold, annotator passes)
- [Cross-dictionary aligned senses, PWG · MW · Apte · MD · ŚKDR · VCP (H3744 slice 1 + H3862 slice 2 + H4751 W2 scaling + H4745 MD extension + H5273 root-vs-nominal gate)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 67330 rows; format=tsv
- [Per-sense corpus attestation (नागदन्त layer, H1455 wave-1)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 87092 rows; format=tsv
- [Pilot cross-dictionary sense view (PWG · MW · Apte) — 500 headwords](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 7359 rows; format=tsv
- [Per-sense first-attestation era buckets (PWG v1, H4019, additive)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 7349 rows; format=tsv
- [Diachronic sense portraits — per-sense frequency x first-era buckets (H4735 join)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 7349 rows; format=tsv
- [Śiva Sūtras (Māheśvara Sūtras) + pratyāhāra table, machine-readable](https://github.com/gasyoun/kosha): MIT (Pāṇini's grammar itself is public domain); 42 rows; format=tsv+json
- [DCS parallel-passage alignments, 'stop-word' variant export (PARA/Stopovye)](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; unspecified rows; format=csv
- [Beginner subhāṣita reader pack — graded Indische Sprüche band (W-RU-b + gloss.ru)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 106 rows; format=json
- [csl-santam Tamil/Capeller/MW corpus fold (mwd+cap+otl, Pahlavi separate)](https://github.com/gasyoun/csl-santam): CC BY-NC-SA 3.0 (Cologne scan data, csl-santam LICENSE.md); 325838 rows; format=sqlite (regenerable fold)
- [Thematic vocabulary drills — recognition/recall items grouped by Amarakosa theme](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 5922 rows; format=json
- [Thematic vocabulary axis — corpus vocabulary grouped by Amarakosa varga](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 2961 rows; format=tsv
- [Cross-dictionary union headword index](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 323422 rows; format=tsv; doi:10.5281/zenodo.22102090
- [Cross-dictionary union headword index, 16 dicts (now-2026 key1+key2)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 417184 rows; format=tsv
- [Unified upasarga atlas — prefixed-verb decomposition x Gita semantics x verb-class frequencies](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 13436 rows; format=tsv
- [MW uttarapada index × DCS Kompozity — dictionary productivity vs corpus attestation](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 19177 rows; format=tsv
- [Consonant series (varga) by DCS period + Cramér's V](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 5 rows; format=csv
- [vidyut-chandas metre definitions (vendored)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 144 rows; format=tsv
- [Frequency-graded vocabulary curriculum — ordered teaching syllabus](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 6667 rows; format=tsv
- [Vocabulary drills — recognition/recall practice items](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 13334 rows; format=json
- [Which-dictionary routing shared-task benchmark (v1.0)](https://github.com/sanskrit-lexicon/csl-guides): CC BY-SA 4.0; 24 rows; format=json
- [Wisdomlib L8 English definitions × CDSL headwords crosswalk (definition → headword linkset)](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 41697 rows; format=tsv.gz
- [Layered Sanskrit corpus extracts of the wisdomlib.org Sanskrit portal (full store, L3-L8, 2026)](https://www.wisdomlib.org/sanskrit): CC BY 4.0; 593587 rows; format=pack.zip (168 entries: tsv/jsonl/txt/md/cff/LICENSE-DATA + MANIFEST.json sha256 per file) + 4 loose root docs on the record; doi:10.5281/zenodo.22117832
- [Zaliznyak declension-class drills — classify a headword or spot the odd one out](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 3434 rows; format=json
- [Zaliznyak-style compact grammar-token index over all PWG](https://github.com/gasyoun/SanskritLexicography): CC BY-SA 4.0; 98639 rows; format=tsv; doi:10.5281/zenodo.22102090
- [А.А.Зализняк public-lecture transcript census (Sanskrit/Vedic + Slavic linguistics)](https://github.com/gasyoun/SanskritLexicography): unspecified; 272 rows; format=tsv (index only — full transcripts NOT vendored, source is a private Yandex.Disk remote)
- [Zaliznyak -ant/-at oddball drills — split-verdict exception bank](https://github.com/gasyoun/VisualDCS): CC BY-SA 4.0; 1722 rows; format=json
- [Zaliznyak token distribution over 98,639 headwords (H6053)](https://github.com/gasyoun/kosha): CC BY-SA 4.0; 342 rows; format=tsv + json + svg

Catalog home: https://github.com/gasyoun/kosha — dataset manifest: data/manifest/datasets.json