Skip to content
Back to search

turkish-native

Agent skill for writing native Turkish and removing translationese from AI and localized copy.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/oguzhankayan--turkish-native/badge.svg)](/listings/oguzhankayan--turkish-native)

HTML

<a href="/listings/oguzhankayan--turkish-native"><img src="/api/listings/oguzhankayan--turkish-native/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Skills CLI

npx skills add oguzhankayan/turkish-native --global

Claude Code plugin

/plugin marketplace add oguzhankayan/turkish-native
/plugin install turkish-native@turkish-native

Manual

Copy `SKILL.md` and the bundled `references/` directory into the skill directory used by your agent.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 18 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "$schema": "https://json.schemastore.org/claude-code-marketplace-manifest.json",
  3. "name": "turkish-native",
  4. "owner": {
  5. "name": "Oğuzhan Kayan",
  6. "url": "https://github.com/oguzhankayan"
  7. },
  8. "description": "Install Turkish Native as a Claude plugin.",
  9. "plugins": [
  10. {
  11. "name": "turkish-native",
  12. "source": "./",
  13. "description": "Write Turkish as native Turkish instead of translated English.",
  14. "license": "MIT",
  15. "keywords": ["turkish", "writing", "localization", "translationese", "copywriting"]
  16. }
  17. ]
  18. }
README.md · README · 123 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Turkish Native
  2. **Write Turkish as Turkish, not translated English.**
  3. [![Validate package](https://github.com/oguzhankayan/turkish-native/actions/workflows/validate.yml/badge.svg)](https://github.com/oguzhankayan/turkish-native/actions/workflows/validate.yml) [![Release](https://img.shields.io/github/v/release/oguzhankayan/turkish-native?display_name=tag)](https://github.com/oguzhankayan/turkish-native/releases/latest) [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
  4. [Türkçe README](README.tr.md) · [Skill](SKILL.md) · [Examples](examples/before-after.md) · [Release](https://github.com/oguzhankayan/turkish-native/releases/latest) · [Contributing](CONTRIBUTING.md) · [Roadmap](ROADMAP.md)
  5. Turkish Native is an open-source agent skill for writing and editing natural Turkish. It targets a specific failure mode in AI-generated and localized copy: Turkish that is grammatically understandable but still carries foreign sentence structure, unnatural collocations, literal SaaS or sales calques, internal product jargon, or model-written marketing formulas.
  6. A sentence can be grammatically correct and still sound translated.
  7. ## Why this exists
  8. Spell checkers and grammar tools catch mechanical errors. Turkish Native works one layer above them: **native construction**. It asks an agent to rebuild the sentence from meaning instead of preserving the source-language skeleton.
  9. That means checking syntax before vocabulary: subordinate clauses, nominalization, word order, case suffixes, possessives, verb–noun pairings, product language, register, factual fidelity, and page-level domain consistency.
  10. ## Examples
  11. | Artificial / translated | Natural Turkish |
  12. |---|---|
  13. | `Kim açtı, kim düzenledi gör.` | `Kimin açtığını, kimin düzenlediğini gör.` |
  14. | `Sayfayı WhatsApp'ta gönder.` | `Sayfayı WhatsApp'tan gönder.` |
  15. | `16 işletme bulundu, 12'si telefonlu.` | `16 işletme bulundu, 12 telefon numarası bulundu.` |
  16. | `Aynı işletme için tekrar kayıt oluşturulmaz.` | `Aynı işletme ikinci kez eklenmez.` |
  17. | `Kart istemiyoruz.` | `Kredi kartı gerekmez.` |
  18. | `Buket için randevu almak istiyorum.` | `Buket siparişi vermek istiyorum.` |
  19. The skill also protects against overcorrection. Natural phrases such as `Sen satışa bak.`, `Bulamazsa uydurmaz.`, `Site kurmak`, or `Linki gönder.` should stay natural when the context supports them.
  20. ## Scope
  21. Use Turkish Native for marketing, product/UI copy, SaaS, localization, documentation, support, legal/privacy/security prose, institutional communication, generated business pages, and reviews of AI-written Turkish.
  22. It is not a spell checker, AI detector, loanword ban list, or replacement for professional legal review. It does not claim that Turkish has only one correct phrasing.
  23. ## Installation
  24. ### Skills CLI
  25. ```bash
  26. npx skills add oguzhankayan/turkish-native --global
  27. ```
  28. ### Claude Code plugin
  29. ```text
  30. /plugin marketplace add oguzhankayan/turkish-native
  31. /plugin install turkish-native@turkish-native
  32. ```
  33. ### Manual
  34. Copy `SKILL.md` and the bundled `references/` directory into the skill directory used by your agent.
  35. ## Usage
  36. ```text
  37. Use $turkish-native to review this Turkish copy.
  38. Flag translationese and unnatural collocations, then rewrite it naturally.
  39. ```
  40. ```text
  41. Translate this into native Turkish with $turkish-native.
  42. Preserve facts and product terminology, but do not preserve English sentence structure.
  43. ```
  44. ```text
  45. Use $turkish-native to rewrite the prose in src/content/tr/home.ts.
  46. Keep keys, variables, URLs, code, and route names unchanged.
  47. ```
  48. ## How it works
  49. Turkish Native uses a meaning-first workflow: identify the audience and register, separate facts from wording, fix structural problems before isolated vocabulary, rebuild foreign sentence architecture, then check native flow, factual fidelity, domain fit, and over-editing.
  50. The taxonomy contains **57 patterns** in six groups:
  51. - 1–10: sentence architecture,
  52. - 11–20: verb choice and lexical fit,
  53. - 21–30: product, UI, and localization,
  54. - 31–40: marketing and model-written copy,
  55. - 41–51: register, fidelity, and final checks,
  56. - 52–57: domain and page integrity.
  57. Detailed rules, examples, and false-positive guards live in [`references/`](references/). The compact runtime instructions live in [`SKILL.md`](SKILL.md).
  58. ## Evaluation
  59. The repository ships with **105 regression cases** across [`evals/cases/`](evals/cases/), including positive controls that should remain unchanged. Every numbered pattern has evaluation coverage.
  60. ```bash
  61. python3 -m pip install -r requirements-dev.txt
  62. python3 scripts/validate-package.py
  63. python3 scripts/validate-benchmark.py
  64. ```
  65. The package validator checks package structure, continuous pattern numbering,
  66. version consistency, unique case IDs, positive-control count, and per-pattern
  67. eval coverage. The benchmark validator separately checks the held-out corpus
  68. schema, unique inputs, scoring dimensions, and pattern/group associations. For
  69. benchmark methodology, see [`docs/BENCHMARKING.md`](docs/BENCHMARKING.md).
  70. ## Project status
  71. Current release: **v0.5.0, pre-1.0**. Pattern names and the evaluation schema may still change before 1.0.
  72. See [`docs/PHILOSOPHY.md`](docs/PHILOSOPHY.md) and [`ROADMAP.md`](ROADMAP.md) for design rationale and next steps.
  73. ## Relationship to Humanizer
  74. [blader/humanizer](https://github.com/blader/humanizer) is a general-purpose skill for broad AI-writing patterns. Turkish Native solves a narrower Turkish-specific problem: translationese and non-native construction.
  75. Turkish Native is not a fork and does not reuse Humanizer's pattern taxonomy. Humanizer inspired the packaging approach.
  76. ## Contributing
  77. Contributions are welcome. Read [`CONTRIBUTING.md`](CONTRIBUTING.md) before opening a PR.
  78. ## License
  79. MIT © 2026 Oğuzhan Kayan. See [`LICENSE`](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.