Marathi transliterator
Skill tejasashinde/marathi-agent-skills/skills/marathi-transliterator
A collection of modular AI agent skills for native Marathi language processing with Claude Code, Cline, etc
npx -y skills add tejasashinde/marathi-agent-skills --skill marathi-transliteratorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Transliterate Marathi between Devanagari and Roman script only. Use when the user explicitly asks for Marathi transliteration or script conversion. Do not use for translation.
SKILL.md
4.7 KB, as published. Nobody here has run it
Marathi Transliterator
Performs deterministic script conversion between Marathi (Devanagari) and Roman letters.
You are performing script transliteration only. If the user provides Marathi text, convert characters to Roman letters using the defined phonemic mapping. If the user provides Romanized Marathi, convert it into valid Devanagari script. If ambiguity exists - ask for clarification. Do not guess.
Do NOT translate meaning between Marathi and English. Words must remain identical phonetically. Only the writing system changes. Example: राम -> rām NOT: hello -> नमस्कार If the input is clearly English language content rather than Romanized Marathi, ask the user to confirm the intended direction.
When to Use This Skill
Load this skill when user says:
- “Transliterate Marathi to Roman letters.”
- “Convert Roman to Marathi script.”
- “Write this Marathi in Latin alphabet.”
- “Give Marathi in Roman letters.”
- “Convert this Roman Marathi into Devanagari.”
Do NOT load this skill for:
- English to Marathi translation
- Marathi to English translation
- Grammar correction or meaning explanation
- Meaning translation
- Paraphrasing.
- Language explanation.
WORKFLOW
1. Detect Script Direction
- If Input contains Devanagari characters (e.g., म, क, आ, ण) -> Convert to Roman.
- If input contains Latin letters that resemble Romanized Marathi phonetics (e.g., aa, ā, bh, dh, kṣ), convert to Marathi.
- If the text is clearly English language content, do NOT transliterate and ask the user for confirmation.
- If mixed or ambiguous -> Ask user to confirm intended direction if not specified.
No assumptions.
2. Apply Deterministic Mapping
Use the exact phonemic mapping below.
Marathi to Roman Transliteration Mapping
1. Independent Vowels
अ -> a आ -> ā इ -> i ई -> ī उ -> u ऊ -> ū ए -> e ऐ -> ai ओ -> o औ -> au अं -> aṃ अः -> aḥ
2. Consonants
क -> k ख -> kh ग -> g घ -> gh ङ -> ṅ च -> c छ -> ch ज -> j झ -> jh ञ -> ñ ट -> ṭ ठ -> ṭh ड -> ḍ ढ -> ḍh ण -> ṇ त -> t थ -> th द -> d ध -> dh न -> n प -> p फ -> ph ब -> b भ -> bh म -> m य -> y र -> r ल -> l व -> v श -> ś ष -> ṣ स -> s ह -> h
3. Dependent Vowel Signs (Matras)
ा -> ā (का -> kā) ि -> i (कि -> ki) ी -> ī (की -> kī) ु -> u (कु -> ku) ू -> ū (कू -> kū) े -> e (के -> ke) ै -> ai (कै -> kai) ो -> o (को -> ko) ौ -> au (कौ -> kau) ं -> ṃ (कं -> kaṃ) ः -> ḥ (कः -> kaḥ)
4. Special Conjuncts
क्ष -> kṣ ज्ञ -> jñ त्र -> tr श्र -> śr
3. Reverse Mapping Rules (Roman -> Marathi)
When converting Roman to Devanagari:
- Recognize digraphs: kh, gh, ch, jh, ṭh, ḍh, th, dh, ph, bh.
- Recognize clusters: kṣ, jñ, tr, śr.
- Treat ā, ī, ū, ṇ, ṭ, ḍ, ś, ṣ as atomic phonemes.
- Default inherent vowel = अ when consonant appears without vowel marker.
- If multiple valid spellings exist, ask for clarification instead of guessing.
- Preserve punctuation, spacing, numerals, and symbols exactly unless the script conversion itself requires a change.
4. Preserve Structure
- Preserve punctuation exactly.
- Preserve spacing.
- Do not alter capitalization except where required by script rules.
- Do not add or remove words.
- Never replace Roman words with Marathi synonyms.
- Never correct grammar.
- Never change word order.
- Never normalize a phonetic spelling into a different Marathi word form.
Validation Rules
Before returning output:
- Ensure no orphaned matras.
- Ensure conjuncts are valid Devanagari combinations.
- Ensure Roman output uses correct diacritics.
- If input is malformed, request correction.
- If the text mixes English and Marathi but the dominant content is not Romanized Marathi, ask before converting.
Examples
Marathi -> Roman
मराठी भाषा सुंदर आहे -> marāṭhī bhāṣā sundar āhe.
Roman -> Marathi
namaskār -> नमस्कार
Conjunct Handling
क्षत्रिय -> kṣatriya
Enforcement Principle
This skill performs deterministic script conversion.
Allowed: मराठी -> marāṭhī bhāṣā -> भाषा
NOT ALLOWED: hello -> नमस्कार good morning -> शुभ सकाळ