# Localize Cs

> Use when translating or localizing UI strings, marketing copy, documentation, or any source text into Czech (cs). Czech is NOT Slovak — enforces Czech vocabulary (soubor, uživatel, složka), letters ě ř ů, 7-case grammar with vocative, 3 genders with animate/inanimate split, ICU one/few/other plurals, perfective/imperfective aspect, Vy/ty formality, impersonal status voice (never first-person Počí…

- **Type:** Skill
- **Install:** `agentstack add skill-i18n-agent-i18nstack-localize-cs`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [i18n-agent](https://agentstack.voostack.com/s/i18n-agent)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [i18n-agent](https://github.com/i18n-agent)
- **Source:** https://github.com/i18n-agent/i18nstack/tree/main/localize-cs
- **Website:** https://i18nagent.ai

## Install

```sh
agentstack add skill-i18n-agent-i18nstack-localize-cs
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Translate to Czech (cs) — High-Fidelity Skill

## Scope & Variants

Czech is a single standard target — Modern Standard Czech (spisovná čeština). No regional split for product UI:

| Locale | Notes |
|--------|-------|
| `cs` / `cs-CZ` | Standard Czech (Czechia). Default and only target. |

**Note on "Czech Republic" vs "Czechia":** the country's official short name is now **Česko / Czechia** (adopted 2016). Use the short form in UI; reserve `Česká republika` for formal/legal contexts.

**Practical reality:** Czech translation has one target. The defining quality concern is **disambiguating from Slovak**, which is structurally similar but lexically distinct.

### Czech vs Slovak — distinct West Slavic languages, often confused

Czech and Slovak are mutually intelligible (especially in writing) but they are **separate languages** with distinct vocabulary, distinct alphabets, and distinct grammar in places. This skill is the mirror partner to `localize-sk`.

**Czech has 3 letters Slovak doesn't:**

| Czech | Used for | Slovak equivalent |
|-------|----------|-------------------|
| **ě** | /je/ or palatalizing previous consonant | not used (Slovak uses `e` or different forms) |
| **ř** | /rž/ (uniquely Czech phoneme) | not used (Slovak has plain `r`) |
| **ů** | long /u:/ (alternant of `ó` from morphological change) | uses `ú` instead |

**Slovak has 5 letters Czech doesn't:** `ľ ĺ ŕ ô ä`. Any of these appearing in Czech text means a Slovak word leaked in.

**Common Czech vs Slovak vocabulary** (the surface markers a translation must NOT mix):

| English | ✓ Czech | ✗ Slovak (do NOT use in cs) |
|---------|---------|-----------------------------|
| file | soubor | súbor |
| folder | složka | zložka / priečinok |
| user | uživatel | používateľ |
| settings | nastavení | nastavenia |
| browser | prohlížeč | prehliadač |
| but | ale | ale (same) |
| because | protože | pretože |
| yes | ano | áno (with diacritic) |
| no | ne | nie |
| month | měsíc | mesiac |
| Monday | pondělí | pondelok |
| Tuesday | úterý | utorok |
| only | jen / pouze | len / iba |
| also | také / taky | tiež / aj |
| can | moci / moct | môcť |
| want | chtít | chcieť |
| see | vidět | vidieť |
| do | dělat | robiť |
| big | velký | veľký |

**If a Czech translation contains `súbor`, `používateľ`, `nastavenia`, `prehliadač`, `áno`, `mesiac`, `pondelok`, `robiť`, `vidieť`, or any of the letters `ľ ĺ ŕ ô ä`** — IT'S WRONG (Slovak leaked in). Fix to the Czech equivalents.

---

## Register Auto-Detection (Apply Before Translating)

Czech has a strong T-V distinction (Vy/ty — called "vykání" vs "tykání"). Auto-detect from source:

| Source signal | Target register |
|---------------|-----------------|
| Casual / conversational / personal (`Hey!`, contractions, emoji, marketing voice, gaming, social) | **Informal (ty / tykání)** — `ty` lowercase, 2nd-person sg. verb endings (`-š`: můžeš, vidíš), familiar imperative (singular: vyber, klikni, otevři). |
| Neutral product UI / docs / consumer software (DEFAULT) | **Formal (Vy / vykání)** — `Vy` capitalized for direct address, 2nd-person pl. verb endings (`-te`: můžete, vidíte), polite imperative (`-te/-ete`: vyberte, klikněte, otevřete). Possessive `Váš/Vaše` capitalized. |
| Legal / banking / government / enterprise B2B | **Formal (Vy) elevated** — same `Vy` form but full constructions, no contractions, prefer impersonal reflexive (`Provádí se`), native vocabulary. |

**Hard rule: never mix levels in one text.** A string with `Vy můžete` (formal verb) and `tvoje nastavení` (informal possessive) is broken.

**Capitalization rule for Vy:** `Vy / Váš / Vaše / Vám / Vás` are capitalized in direct second-person address.

**Default for software UI: Vy (formal)** unless brand voice is explicitly youth/casual.

---

## Critical Hard Rules (Severity ≥ 2.0 — Auto-Fail If Wrong)

### 1. Czech identity — NOT Slovak (severity 2.5)

Documented above. Check for Slovak letters (`ľ ĺ ŕ ô ä`) and Slovak vocabulary.

### 2. Word integrity — verb prefixes NEVER split (severity 3.0)

Czech verbs with prefixes are **single words**. Splitting them is a critical error.

| ✗ Wrong (split) | ✓ Correct (joined) |
|-----------------|---------------------|
| `pře vést` | **`převést`** (to transfer) |
| `pro vést` | **`provést`** (to execute) |
| `vy konat` | **`vykonat`** (to perform) |
| `na hrát` | **`nahrát`** (to upload / record) |
| `pře nést` | **`přenést`** (to transfer) |
| `od eslat` | **`odeslat`** (to send out) |
| `při jmout` | **`přijmout`** (to accept) |
| `za hájit` | **`zahájit`** (to begin) |

**Reflexive `se / si` is SEPARATE clitic** — not attached to verb in writing. `přihlásit se`, `registrovat se`, `zabývat se`. Writing `přihlásitse` is wrong.

### 3. Czech diacritics — preserve all (severity 2.5)

Czech uses: `á č ď é ě í ň ó ř š ť ú ů ý ž`. Stripping any is a critical readability failure. Especially:
- **ě** (`mě`, `tě`, `pět`, `dělat`) — palatalizing e.
- **ř** (`říjen`, `řeka`, `řekl`) — uniquely Czech sound.
- **ů** (`dům`, `stůl`, `obrátit`) — long u with marker.

| ✗ Stripped | ✓ Correct |
|------------|-----------|
| uzivatel | **uživatel** |
| nastaveni | **nastavení** |
| prohlizec | **prohlížeč** |
| reka | **řeka** |
| dum | **dům** |

### 4. Seven-case system — Czech retains active vocative (severity 2.5)

Czech uses **all 7 Slavic cases**, INCLUDING the vocative (unlike Slovak which merged it into nominative). Every noun, adjective, and pronoun declines.

| Case | Question | Use | Example (m. anim. "uživatel") | Example (f. "aplikace") |
|------|----------|-----|-------------------------------|--------------------------|
| Nominativ (NOM) | kdo? co? | Subject | uživatel | aplikace |
| Genitiv (GEN) | koho? čeho? | Possession, "of", absence, 5+ count | uživatele | aplikace |
| Dativ (DAT) | komu? čemu? | Indirect object, "to" | uživateli | aplikaci |
| Akuzativ (ACC) | koho? co? | Direct object | uživatele (animate=GEN) / soubor (inanimate=NOM) | aplikaci |
| **Vokativ (VOC)** | — | **Direct address — ACTIVE in Czech** | uživateli! | aplikace! |
| Lokál (LOK) | o kom? čem? | Location (always w/ prep) | o uživateli | o aplikaci |
| Instrumentál (INS) | s kým? čím? | "With", means | s uživatelem | s aplikací |

**Critical: animate masculine accusative = genitive form.** `vidím uživatele`, NOT `vidím uživatel`.

**Vocative usage** — Czech ACTIVELY uses vocative for direct address (unlike Slovak/Russian where it's archaic/merged). When a Czech product UI greets the user by name or addresses a role:

| ✗ Wrong (nominative) | ✓ Correct (vocative) |
|----------------------|----------------------|
| Ahoj, Petr! | **Ahoj, Petře!** |
| Vážený uživatel! | **Vážený uživateli!** |
| Děkujeme, Pavel! | **Děkujeme, Pavle!** |
| Karel! | **Karle!** |

Common vocative endings: m.anim. `-e/-i` (Petr → Petře, Karel → Karle, učitel → učiteli), f. usually = nominative (Marie → Marie), n. = nominative.

### 5. Preposition + governed case (severity 2.5)

| Preposition | Case | Example |
|-------------|------|---------|
| v / ve (in) | LOK (location) | v systému |
| do (into) | GEN | do systému |
| na (on/onto) | LOK / ACC | na obrazovce / na obrazovku |
| s / se (with) | INS | s uživatelem |
| bez (without) | GEN | bez chyb |
| pro (for) | ACC | pro uživatele |
| od (from) | GEN | od uživatele |
| k / ke (to/toward) | DAT | k uživateli |
| o (about) | LOK | o uživateli |
| po (after) | LOK | po přihlášení |
| před (before) | INS / ACC | před uložením |
| za (per / behind) | ACC / INS | za soubor / za stolem |
| u (at, by) | GEN | u uživatele |
| z / ze (from) | GEN | z aplikace |

**"per" in Czech:** use `za + ACC` for rate (`za soubor`), `na + ACC` for distribution (`na uživatele`), `denně` for "per day". Never `per soubor` (English calque).

### 6. ICU plurals — one / few / other (1 / 2-4 / 0+5+ with 11-14 going to other) (severity 2.5)

```icu
{count, plural,
  one {# soubor}
  few {# soubory}
  other {# souborů}
}
```

**CLDR plural category boundaries:**

| Category | Rule | Examples | Noun form |
|----------|------|----------|-----------|
| `one` | n = 1 | 1 | **NOM singular** (soubor) |
| `few` | n ∈ {2, 3, 4} | 2, 3, 4 | **NOM plural** (soubory) |
| `other` | n = 0 OR n ≥ 5 OR fractions | 0, 5-100, 1.5, 11, 12, 13, 14 | **GEN plural** (souborů) |

**Note vs other Slavic:** Czech is **simpler than Russian/Ukrainian/Croatian/Slovak** here — only `one / few / other` (3 categories), no `many` category. 11-14 fall into `other`, NOT `few`.

### 7. Numeral-noun agreement (severity 2.0)

| Count | Pattern | Example |
|-------|---------|---------|
| 1 | NOM sg | 1 soubor |
| 2, 3, 4 | NOM pl | 2 soubory, 3 systémy |
| 5+ | GEN pl | 5 souborů, 10 uživatelů |
| 11-14 | GEN pl (same as other 5+) | 11 souborů, 12 systémů |
| 21 | NOM sg (back to "one") | 21 souborů — WAIT, actually Czech is different from Russian here |
| 22-24 | varies by interpretation — commonly NOM pl when "21 = 1+20" interpretation isn't applied | (Czech generally treats 21+ as `other` = gen.pl. in modern usage) |

**Modern Czech tends to use GEN pl for everything 5+** including 21, 22, etc., though strict prescriptive grammar sometimes calls for NOM sg with 21, 22 etc. For UI, GEN pl is the safer default for 5+.

### 8. Three-gender + masculine animate/inanimate (severity 2.0)

Czech distinguishes **masculine animate** vs **masculine inanimate** (matters for accusative case):

| Gender | Example | Adjective NOM sg. |
|--------|---------|-------------------|
| Masc. animate (humans, named animals) | uživatel, programátor | nový uživatel |
| Masc. inanimate (objects, abstracts) | soubor, systém, počítač | nový soubor |
| Feminine (-a ending mostly) | aplikace, složka, databáze | nová aplikace |
| Neuter (-o/-e/-í ending) | nastavení, okno, zařízení | nové nastavení |

Wrong gender adjective is auto-fail: `nová soubor` ✗, `nový aplikace` ✗, `nová nastavení` ✗.

### 9. Past-tense gender agreement (severity 1.5)

| Subject | L-participle | Example |
|---------|--------------|---------|
| m. sg. | -l | on pracoval |
| f. sg. | -la | ona pracovala |
| n. sg. | -lo | ono pracovalo |
| m. anim. pl. | -li | oni pracovali |
| f. pl. / m. inan. pl. | -ly | ony pracovaly / soubory se uložily |
| n. pl. | -la | okna se otevřela |

### 10. Clitic second-position rule (severity 2.0)

Czech clitics (`se`, `si`, `by`, `bych`, `bys`, `jsem`, `jsi`, `mi`, `ti`, `mu`, `jí`, `ho`, `tě`, etc.) MUST go in the **second position** of their clause (Wackernagel's law), same as other Slavic languages.

| ✗ Wrong | ✓ Correct |
|---------|-----------|
| `Se mi to nelíbí` | **`To se mi nelíbí`** |
| `Soubor se mu zobrazuje` | **`Soubor se mu zobrazuje`** ✓ — wait, this is correct because `se mu` cluster is in position 2. The issue would be: `Soubor mu se zobrazuje` ✗ → `Soubor se mu zobrazuje` ✓ (clitic order: refl. `se` after AUX, before DAT) — actually Czech clitic order is: AUX → DAT → ACC → SE. So `se mu` is wrong; correct is `mu se`. Let me revise: |
| `Soubor se mu zobrazuje` (if mu comes first lexically) | Actually `Soubor mu se zobrazuje` ← but in Czech `se` typically comes after dative? Let me use a clear example. |

**Clitic cluster order in Czech:**
1. AUX (`jsem / jsi / je / jsme / jste / jsou / bych / bys / by / bychom / byste`)
2. SE / SI (reflexive)
3. DAT (`mi / ti / mu / nám / vám / jim`)
4. ACC (`mě / tě / ho / nás / vás / je`)

Example: `Já bych se mu byl zeptal.` (AUX bych — SE se — DAT mu — past). For UI status messages this complex cluster is rare; the main rule is: **`se / si` stays in 2nd position of the clause**.

### 11. Verb aspect — perfective vs imperfective (severity 2.0)

Standard UI verb aspect pairs:

| Imperfective (process, ongoing) | Perfective (single completed event) | English |
|--------------------------------|--------------------------------------|---------|
| ukládat | uložit | save |
| mazat | smazat | delete |
| nahrávat | nahrát | upload / record |
| stahovat | stáhnout | download |
| otevírat | otevřít | open |
| zavírat | zavřít | close |
| vytvářet | vytvořit | create |
| upravovat | upravit | edit |
| měnit | změnit | change |
| posílat | poslat | send |
| psát | napsat | write |
| číst | přečíst | read |
| vybírat | vybrat | choose / select |
| překládat | přeložit | translate |
| přihlašovat se | přihlásit se | log in |

**UI patterns by aspect:**
- Buttons (single action) → **perfective infinitive**: `Uložit`, `Smazat`, `Poslat`, `Vybrat`. The Czech UI convention is INFINITIVE (not imperative).
- Ongoing status → **imperfective verbal noun** (`Ukládání…`, `Načítání…`) or **`Probíhá + verbal noun`**.
- Completed → **perfective neuter participle** (`Uloženo`, `Načteno`, `Přeloženo`).

### 12. Verb-construction integrity — no stacked finite verbs (severity 2.0)

English `-ing` (gerund as adverbial) MUST be rendered with **přechodník** (Czech verbal adverb in `-ouc/-íc/-e`) OR restructure with `při + verbal noun`. Note: the přechodník is somewhat literary/archaic in modern Czech — `při + verbal noun` is more common in product copy.

| ✗ Stacked finite | ✓ Restructured | English |
|------------------|-----------------|---------|
| `Strávil jsem 6 let stavěl jsem firmu` | **`Strávil jsem 6 let stavbou firmy`** OR **`Strávil jsem 6 let při budování firmy`** | I spent 6 years building the business |
| `Pracoval jsem dokud jsem psal zprávu` | **`Pracoval jsem při psaní zprávy`** | I worked while writing the report |

---

## Pronouns and Possessives

### Personal pronouns

| English | Czech | Notes |
|---------|-------|-------|
| I | já | |
| you (sg. informal) | ty | |
| you (sg./pl. formal — capitalize) | Vy | Always capitalized in direct address |
| you (pl. informal) | vy | lowercase when truly plural informal |
| he | on | |
| she | ona | |
| it | ono | |
| we | my | |
| they (m. anim.) | oni | |
| they (m. inan. / f.) | ony | |
| they (n.) | ona | |

### Possessive pronouns

| Person | m. sg. | f. sg. | n. sg. |
|--------|--------|--------|--------|
| můj (my) | můj | moje / má | moje / mé |
| tvůj (your sg.) | tvůj | tvoje / tvá | tvoje / tvé |
| **Váš (your formal — capitalize)** | **Váš** | **Vaše / Vaše** | **Vaše** |
| jeho (his — invariable) | jeho | jeho | jeho |
| její (her) | její | její | její |
| svůj (reflexive — own) | svůj | svoje / svá | svoje / své |
| náš (our) | náš | naše | naše |
| jejich (their — invariable) | jejich | jejich | jejich |

**Reflexive `svůj`** is required when possessor = subject of same clause.

---

## UI Conventions

### Buttons — INFINITIVE (the Czech UI convention)

Czech platform UIs (Windows CZ, Chrome CZ, macOS CZ) use **infinitive** for buttons — NOT imperative. This is a deliberate convention choice and matches what users see in their OS.

| English | ✓ Infinitive (button) | (Vy-imperative — for instruction prose) |
|---------|------------------------|------------------------------------------|
| Save | **Uložit** | Uložte |
| Cancel | **Zrušit** / **Storno** | Zrušte |
| Delete | **Smazat** / **Odstranit** | Smažte / Odstraňte |
| Send | **Odeslat** | Odešlete |
| Edit | **Upravit** | Upravte |
| Search | **Hledat** | Hledejte |
| Confirm | **Potvrdit** | Potvrďte |
| Continue | **Pokračovat** | Pokračujte |
| Submit | **Odeslat** | Odešlete |
| Sign in / Log in | **Přihlásit se** | Přihlaste se |
| Sign out | **Odhlásit se** | Odhlaste se |
| Sign up | **Zaregistrovat se** | Zaregistrujte se |
| Next | **Další** / **Pokračovat** | (same) |
| Back | **Zpět** | (same) |
| Done | **Hotovo** / **Dokončeno** | (same) |
| OK | **OK** / **Souhlasím** | (same) |
| Close | **Zavřít** | Zavřete |
| Upload | **Nahrát** | Nahrajte |
| Download | **Stáhnout** | Stáhněte |
| Refresh | **Aktualizovat** / **Obnovit** | Aktualizujte |
| Settings | **Nastavení** | (same) |
| Apply | **Použít** | Použijte |
| Reset | **Resetovat** / **Obnovit původní** | — |
| Select all | **Vybrat vše** | Vyberte vše |
| Deselect all | **Odznačit vše** / **Zrušit vše** | — |
| Add more | **Přidat dal

…

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [i18n-agent](https://github.com/i18n-agent)
- **Source:** [i18n-agent/i18nstack](https://github.com/i18n-agent/i18nstack)
- **License:** MIT
- **Homepage:** https://i18nagent.ai

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-i18n-agent-i18nstack-localize-cs
- Seller: https://agentstack.voostack.com/s/i18n-agent
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.