Persian Language Tech skills for AI agents

Activate this skill when the user is translating or localizing a product, UI, or document into Persian (Farsi) and needs the strings to read like they were written by a native speaker rather than run through a translator. Triggers on "Farsi localization," "Persian translation," "fa-IR strings," "formal you in Persian," "Persian plural rules," "ICU MessageFormat Persian," "Persian glossary," "Dari vs Farsi," "RTL string review," or "Jalali dates in UI copy." Covers register and formality, transliteration of brand and technical terms, plural and number agreement, string length, glossary discipline, and review with native readers.

Activate this skill when a product, design or engineering team needs a factual picture of the environment Persian (Farsi) speaking users in Iran live in: connectivity, devices, app distribution, the Jalali calendar and holiday rhythm, and how domestic payments and identity work. Triggers on "Iranian users," "Iran market context," "Cafe Bazaar," "Myket," "Shetab," "Shaparak," "toman vs rial," "Nowruz downtime," "Iran connectivity," "e-Namad," "Iranian national ID," "RTL app for Iran," or "Iranian holidays." Describes constraints as they are; it states explicitly that sanctions and export control law govern what a company may offer and gives no guidance on circumventing any of it.

Activate this skill when the user needs to store, convert, format, parse or validate dates in the Solar Hijri calendar used in Iran and by Persian (Farsi) speakers, or is debugging a Jalali date bug in a web, mobile or backend system. Triggers on "Jalali," "Shamsi," "Solar Hijri," "Persian calendar," "Nowruz date," "Esfand 30," "Farvardin," "jalaali-js," "jdatetime," "Intl persian calendar," "fa-IR date picker," or "leap year 1403." Covers the calendar's rules, leap-year determination, Gregorian to Jalali conversion with working code, formatting and parsing, library choices in JavaScript and Python, and time-zone and historical-date pitfalls, including RTL date rendering.

Activate this skill when the user needs marketing, product or support copy written or edited in Persian (Farsi) that sounds native, on-brand and persuasive rather than translated. Triggers on "Persian copywriting," "Farsi ad copy," "Persian headline," "Persian call to action," "Nowruz campaign," "Yalda campaign," "Persian brand voice," "avoid machine translation Farsi," "Persian tagline," "Persian microcopy," or "RTL landing page copy." Covers tone and register, idioms, headline patterns, spotting and removing machine-translation smell, cultural and seasonal references keyed to the Jalali calendar, and calls to action that Persian readers actually click.

Activate this skill when the user wants organic search visibility for Persian (Farsi) content or a site aimed at Persian-speaking users in Iran, Afghanistan or the diaspora. Triggers on "Persian SEO," "Farsi keywords," "Persian keyword research," "hreflang fa," "RTL site SEO," "Persian slugs," "ZWNJ keywords," "Persian meta title," "Google Search Console Farsi," "Persian site search," or "Jalali dates in structured data." Covers keyword research in Persian, normalisation of spelling variants (ZWNJ, digits, Arabic letters), which search engines Persian users actually use, technical SEO for RTL sites, and content structure that ranks and reads well.

Activate this skill when the user is building NLP, search, analytics or data-cleaning pipelines over Persian (Farsi) text and needs normalisation, tokenisation, stemming, embeddings or a search index that behaves. Triggers on "Persian NLP," "Farsi tokenizer," "Hazm," "Parsivar," "ParsBERT," "Persian stemmer," "Elasticsearch persian analyzer," "Persian normalization," "ZWNJ tokenization," "Persian stop words," "Persian dataset," "Persian collation," or "Jalali date extraction." Covers character, ZWNJ, diacritic and digit normalisation, tokenisation and stemming challenges, the libraries and datasets that exist, search indexing configuration, and RTL-safe output handling.

Activate this skill when the user is rendering, storing, searching or cleaning Persian (Farsi) text and hits problems with joined letters, the zero-width non-joiner, Arabic versus Persian code points, digit sets, fonts or justification in an RTL layout. Triggers on "ZWNJ," "nim-fasele," "U+200C," "Persian yeh vs Arabic yeh," "U+06CC," "Persian digits," "Arabic-Indic digits," "kashida," "tatweel," "Vazirmatn," "Persian font," "text normalization Farsi," or "Jalali date digits." Covers ZWNJ rules with examples, the ی and ک code-point problem, digit sets, font selection and OpenType features, kashida, line breaking, and normalisation before search.

Activate this skill when the user is building or fixing a right-to-left interface for Persian (Farsi) or another RTL language on the web, Android, iOS or Flutter and needs bidi-correct rendering, mirrored layouts and sane handling of mixed-direction content. Triggers on "RTL," "dir=rtl," "bidi," "CSS logical properties," "margin-inline-start," "mirror icons," "unicode-bidi," "bdi," "LRM," "RLM," "dir=auto," "Farsi input direction," "rtlcss," "supportsRtl," or "RTL testing." Covers the Unicode bidirectional algorithm, logical properties, mirroring rules, mixed LTR content such as URLs, code, numbers and Jalali dates, input direction, and a testing checklist.