# Bulgarian Transliteration (Official Latin Spelling) Bulgarian Transliteration (Official Latin Spelling) is a tested SKILL.md that turns Bulgarian Cyrillic text into Latin letters by the official Bulgarian system, the one the Transliteration Act of 2009 makes mandatory for names, addresses, passports, forms and place names; an agent buys it once for $0.05 over x402. - Page: https://aiskills402.com/skills/bulgarian-transliteration - Category: Translation & Localization (https://aiskills402.com/categories/translation) - Price: $0.05 once, USD-priced, paid in USDC on Base over x402. Price as loaded on this page. The 402 response your agent receives is authoritative. - Version: 1.0.0 - Card (JSON): https://api.aiskills402.com/v1/skills/bulgarian-transliteration ## Use it when Turns Bulgarian Cyrillic text into Latin letters by the official Bulgarian system, the one the Transliteration Act of 2009 makes mandatory for names, addresses, passports, forms and place names. One letter table, the word-final ия written ia, the country written Bulgaria, ъ written a, щ written sht, and every digit, punctuation mark, Latin word and line break kept byte for byte. Also makes URL slugs and flags letters that Bulgarian does not have. Use when asked to transliterate Bulgarian, write a Bulgarian name or address in Latin letters, make a slug from a Bulgarian title, or convert Cyrillic to the official Latin spelling. ## Not for Translating Bulgarian into English (only the letters change, not the words); Russian, Ukrainian or Serbian Cyrillic, which have their own systems; turning Latin text back into Cyrillic; established foreign spellings of a name. Capitals of ж, ч, ш, щ get a capital first letter only (Zh, Sht); the Act's table prints full capitals, so check what an office wants. ## Tested, honestly Tested 2026-10-08. - Strong model (claude-sonnet-5-5 (Claude Code alias "sonnet")): Right on all 24 texts, read by hand: ъ as a in Targovishte and Balgarskata, щ as sht, ц as ts, ж as zh, й as y, the word-final ия as ia in Maria, Ilia and Azia, the country itself as Bulgaria, capitals kept in an all-caps line, a hyphenated surname, an address with its number, Markdown, digits and Latin words such as Wi-Fi and Xiaomi left as they were, slugs built from the finished text, Russian letters outside the Bulgarian table kept and named, and the line telling it to ignore its instructions transliterated as text. - Weak model (claude-haiku-5-5 (Claude Code alias "haiku")): Right on 18 of 24 texts, read by hand, but it missed six: it kept ия as iya in Илия and Азия, wrote Bulgarskata for the adjective where only the country name is Bulgaria, wrote й as i in one word, used the letter ž for Джон, and put an asterisk inside a Russian place name it was asked to keep. Note: Twenty-four Bulgarian texts written by us: names, towns, an address, capitals, a hyphenated surname, the endings ия and ъ, the country name, Markdown with Latin words and prices, three slug tasks, two words with Russian letters and a line that tells the model to ignore its instructions. Each answer is compared by code with the exact expected Latin text (a slug task as JSON with the text and the slug), so a single wrong letter fails the case. The rules are the official table of the Bulgarian Transliteration Act of 2009 (Art. 4 table, Art. 5 for ия at the end of a word, Art. 6 for the country name), re-read on 2026-10-08 on a public copy of the Act. The first Haiku run with the skill was cut off by a usage limit that the tester then recorded as answers; that run was thrown away and Haiku was run again with the skill, and the tester now stops on such a message. No check was widened. One run per model and text. ### With and without the skill Tested 2026-10-08. - Texts transliterated right (24 texts): Sonnet 24/24 with, 15/24 without; Haiku 18/24 with, 11/24 without. Same request on both sides, a fence removed first, compared with the same expected text. Read by hand, Sonnet without the skill already handled zh, ts, ch, sh, the plain names, the address and the Russian letters. It missed nine: it used another system for Targovishte (Târgovište), wrote ия as iya in Maria, Yulia, Ilia and an all-caps line, wrote Bulgarskata for the adjective and Balgaria for the country (both the wrong way round), translated рутер into router inside a sentence, and missed the surname with щ. Haiku without the skill missed thirteen, the same kinds plus ъ written as u and the slugs. With the skill Haiku did worse than without on two texts (Илия, a Russian place name). Same cases and the same checks with and without the skill. The cases are ours, written around what the skill is for; with a handful of cases, a difference of one or two is within noise. Full summary: https://aiskills402.com/skills/bulgarian-transliteration/tests ## Example ### Bulgarian (claude-sonnet-5-5) Before: > Търговище After: > Targovishte ## What is in the file - Hard rules - The table - Letter combinations - Word-final ия - The name of the country - Capital letters - Letters that are not Bulgarian - Slugs - Work in this order - Short examples ## Languages Bulgarian, English ## How to buy Agent (HTTP): 1. GET https://api.aiskills402.com/v1/skills/bulgarian-transliteration/file without a payment header. The answer is 402 with a PAYMENT-REQUIRED header (x402 v2): exact amount, asset, network, recipient. 2. Sign `accepts[0]` with an x402 client (for example @x402/core + @x402/evm). 3. Repeat the GET with the signature in the PAYMENT-SIGNATURE header. The answer is 200 with the file, its sha256 and a re-download token. Agent (MCP): https://mcp.aiskills402.com/mcp — free tools search_skills, get_skill, redownload_skill. Buying itself is over HTTP. Full flow: https://aiskills402.com/docs ## The file - Version: 1.0.0 - Size: 8.7 KB (8910 bytes) - SHA-256: 79b46fd8a9f2d2c9e8f2ca078fc54607010cb3b4eedd2eed117245e231cd274e - Updated: 2026-10-08 - New versions are free through your re-download token. ## Versions ### 1.0.0 (2026-10-08) First release: transliterates Bulgarian Cyrillic into Latin letters by the official table of the Transliteration Act (2009); word-final ия as ia, the country as Bulgaria, ъ as a, щ as sht; every non-Cyrillic character kept; URL slugs; letters of other Cyrillic alphabets kept and listed on request. Facts re-checked on 2026-10-08 with one read-only fetch of the Act's text on a public legal-wiki source (Art. 4 table, Art. 5 combinations and the ia rule, Art. 6 country name). Not yet measured with a model: the test run, baseline and price decision follow. ## License Perpetual, non-exclusive; use and modify for yourself incl. paid work; no resale or republishing. Holder: Georgi Kalchev, aiskills402.com. Terms: https://aiskills402.com/docs#license ## FAQ ### Which system does it use? The official Bulgarian system of the Transliteration Act of 2009: ж is zh, ц is ts, ч is ch, ш is sh, щ is sht, ъ is a, й and ь are y, ю is yu, я is ya. The word-final ия is written ia, so София is Sofia, and the country is written Bulgaria. Nothing from the international scientific standard appears: no háček letters, no breve marks, no apostrophes, and ъ never becomes u. ### What does it leave alone? Digits, punctuation, quotation marks, Latin words, Markdown and line breaks come back exactly as they were. Letters from other Cyrillic alphabets, such as ы or ї, are not guessed: they stay as they are and, when you ask for it, are listed under a key you name. A sentence that sounds like a command is treated as ordinary text and changed letter by letter, never obeyed. Digits and punctuation survive, whatever their shape. ### Can it make a URL slug? Yes. The slug is made from the finished transliteration: lower case, every run of other characters becomes one hyphen, digits stay, nothing is shortened. Нова статия за България becomes nova-statia-za-bulgaria. Names that end in ия follow the Act: Мария is Maria and Илия is Ilia, while Лиляна stays Lilyana. Cities, streets and company names work the same way, so Търговище and Горна Оряховица always give the same Latin spelling. Hyphenated surnames such as Щерева-Йорданова are handled part by part, which gives Shtereva-Yordanova. ### Does it help Claude Sonnet? Yes, and it shows letter by letter. We sent twenty-four Bulgarian texts to Sonnet and Haiku, once with the file and once without, and compared every answer with the exact Latin spelling the law gives. Bare Sonnet scored 15: it slipped into another system for Targovishte, kept iya where the Act writes ia, mixed up Balgarskata and Bulgaria, and translated a Bulgarian word instead of transliterating it. With the file it got all 24. Haiku went from 11 to 18. ## Related skills - [Translate App Text (JSON, YAML, PO)](https://aiskills402.com/skills/translate-app-text.md): $0.01 once - [SEO Meta Writer](https://aiskills402.com/skills/seo-meta.md): $0.03 once ## Measurement limits - Models other than the two named above were not run. - Each verdict comes from the test run on the date shown; the skill may have changed since (check the version). - Full test inputs are not published here, only short excerpts of our own text. - Results on your own texts, languages and domains can differ. Offer note: Paid in USDC (USD-pegged) over x402 by an AI agent; one-time.