magic-linguistic-scripts

When to Use

• Setting normalization policy for a target script (NFC vs NFKC).
• Diagnosing garbled output, unknown glyphs, mojibake.
• Pre-deduplication confusable folding (e.g., Cyrillic "а" vs Latin "a").
• Romanization / transliteration table selection.
• Diacritic restoration or preservation for tone languages.
• Multi-script corpus (mixed Devanagari + Latin Hindi, Cantonese in Han + Latin, Kazakh in Cyrillic + Latin).

When NOT to use: the script policy is already in workspace_state.md and downstream specialists are following it; or the operation is purely byte-level (e.g., gzipping files).

Decision Tree — Pick The Right Operation