Skip to content
Back to search

rontgen

SkillWorks with Claude CodeGitHub

Röntgen — a Claude skill that maps an existing codebase and writes the documents that make it understandable, every sentence tied to a source line. Bilingual EN/TR.

Safety scan

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

Checked today

Check it deeper on your machine

It has changed since its scan on 21 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/XINMurat/rontgen with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/XINMurat/rontgen'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/xinmurat--rontgen/badge.svg)](/listings/xinmurat--rontgen)

HTML

<a href="/listings/xinmurat--rontgen"><img src="/api/listings/xinmurat--rontgen/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

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.

README.md · README · 67 lines
  1. # Röntgen
  2. **Maps an existing codebase and writes the documents that make it understandable — every sentence tied to a source line.**
  3. Part of a family of Claude skills: iskele builds, mizan weighs, kiyas generates, ux-mizan audits experience, **röntgen maps**.
  4. [English](#english) · [Türkçe](#türkçe)
  5. ---
  6. ## English
  7. Hand Röntgen a repository you did not write. It measures first — inventory, hotspots, dependencies resolved through
  8. each language's own imports and package manifests, blind spots, entry points, data stores — and only then reads, in
  9. the order the measurement points to. It writes a reading order, an architecture page (what the system is made of
  10. and who owns which state), module notes, flows, a glossary and a risk list. Every factual sentence cites
  11. `path/file.ext:line`; a second reader who did not write the documents checks a sample before delivery.
  12. **What it is good at, measured** (details in `skill/rontgen/references/kanit.md`): on a 108k-line Rust repository
  13. with its architecture document removed, three runs covered 11.5 of 14 points of that document on average, against
  14. 4.5 to 6 for sessions without the skill, and held 18–19 of 20 sentences in a blind check. On a 139k-line Go
  15. repository one run covered 7 of 14 against 3. **What it is not:** it reads 1–2% of a large codebase; its documents
  16. say so, and name what they did not read.
  17. **Supported:** dependency resolution for Python, TypeScript/JavaScript (relative imports), C#, C/C++, Rust, Go,
  18. Java/Kotlin; manifests for Cargo, npm, .NET, Python, Go, Maven, Gradle; ORM tables for EF Core, SQLAlchemy,
  19. Django, TypeORM, Prisma. Other languages are counted and labelled "unmeasured", never silently zero.
  20. Output: Markdown always; SVG, PNG, Word, LaTeX and PDF when the machine can make them.
  21. ```bash
  22. python skill/rontgen/scripts/rontgen.py --repo . --out rontgen-out
  23. python skill/rontgen/scripts/yapi.py --repo . --out rontgen-out
  24. python skill/rontgen/scripts/paket.py --repo . --out rontgen-out
  25. ```
  26. A real line from `paket.md` on that Rust repository — the tie no import showed:
  27. ```text
  28. | `helix-view` `helix-view/Cargo.toml` | rust | helix-core, helix-dap, … | 24 | helix-tui | crossterm (term), termina (term) |
  29. ```
  30. Standard library only. Every script has `--self-test`. Documentation: [docs/](docs/index.md).
  31. ---
  32. ## Türkçe
  33. Röntgen'e yazmadığınız bir depo verin. Önce ölçer — envanter, sıcak noktalar, her dilin kendi import'ları ve paket
  34. bildirimleri üzerinden çözülen bağımlılıklar, kör noktalar, giriş yüzeyi, veri depoları — sonra ölçümün gösterdiği
  35. sırayla okur. Bir okuma sırası, bir mimari sayfası (sistem neyden yapılmış, hangi durumu kim tutuyor), modül notları,
  36. akışlar, bir sözlük ve bir risk listesi yazar. Her olgusal cümle `yol/dosya.uzanti:satir` gösterir; teslimden önce
  37. belgeleri yazmamış bir ikinci okuyucu bir örneklemi denetler.
  38. **Ölçülen başarısı** (ayrıntı `skill/rontgen/references/kanit.md`): mimari belgesi çıkarılmış 108 bin satırlık bir
  39. Rust deposunda üç koşu, o belgenin 14 maddesinden ortalama 11,5'ini kapsadı; skill'siz oturumlar 4,5–6. Kör denetimde
  40. 20 cümleden 18–19'u tuttu. 139 bin satırlık bir Go deposunda tek koşu 14'te 7 kapsadı (skill'siz 3). **Ne olmadığı:**
  41. büyük bir kod tabanının %1–2'sini okur; belgeleri bunu söyler ve okumadığını adıyla yazar.
  42. **Desteklenen:** Python, TypeScript/JavaScript (göreli import), C#, C/C++, Rust, Go, Java/Kotlin bağımlılık çözümü;
  43. Cargo, npm, .NET, Python, Go, Maven, Gradle paket bildirimleri; EF Core, SQLAlchemy, Django, TypeORM, Prisma tabloları.
  44. Diğer diller sayılır ve "ölçülmedi" diye etiketlenir, sessizce sıfır yazılmaz. Çıktı: her zaman Markdown; makine
  45. üretebiliyorsa SVG, PNG, Word, LaTeX ve PDF.
  46. Yalnız standart kütüphane. Her betiğin `--self-test`'i var. Belgeler: [docs/](docs/index.md).
  47. ---
  48. Code MIT · prose CC BY 4.0.

Ratings and comments

Sign in to rate this skill.

Sign in

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