Humanizer is a .NET library that provides a comprehensive set of extension methods for converting raw data types (numbers, dates, times, byte sizes, strings, enums, collections) into human-readable, localized text. It supports over 90 locales with full pluralization, ordinalization, inflection, and parsing of number words back to numeric values. The library is designed as a reusable dependency for applications requiring user-friendly formatting. Architecturally, Humanizer follows a strategy/registry pattern: each formatting domain (number-to-words, date humanization, ordinalization, etc.) defines an interface and a default implementation, with locale-specific overrides resolved through a central Configurator. Locale data is defined in YAML files and compiled at build time by a Roslyn source generator into immutable profile classes, keeping runtime algorithms generic and data-driven. The library also includes Roslyn analyzers for API migration and a comprehensive test suite with cross-locale theory tests. Key entry points are extension methods on primitive types (e.g., `42.ToWords()`, `"hello_world".Humanize()`, `DateTime.Now.Humanize()`, `ByteSize.FromBytes(1024).Humanize()`) and the `Configurator` for advanced configuration. The primary data flow is: user calls an extension method → culture resolution → delegation to the appropriate converter/formatter → use of generated profile data to produce localized output. External dependencies are limited to the .NET standard library and CLDR plural rule data.
Patina is a deterministic, pattern-based AI text detection and humanization tool, primarily targeting Korean but supporting English, Chinese, and Japanese. It operates as a Node.js CLI (`patina-cli`), a web API (Vercel serverless), and an agent skill (Claude Code, Codex, Cursor). The core workflow: analyze source text for AI-typical patterns (stylometry, lexicon, discourse tells), build a structured prompt, invoke an LLM backend (OpenAI, Claude, Gemini, Kimi, etc.), verify meaning preservation (MPS/fidelity), and output a humanized rewrite with a deterministic audit trail. It enforces strict meaning guards (number safety, polarity, causation) and provides scoring, diff, and verification modes. The architecture is modular: a shared deterministic analysis core (`src/features/`) runs identically in Node and browser; a prompt builder assembles context from pattern packs, document types, and personas; multiple backend adapters abstract LLM invocation; and a verification pipeline ensures fidelity. The project includes extensive research infrastructure (benchmarking, human evaluation panels, corpus management) and a commercial Pro tier with Polar license validation. It is well-tested with over 200 unit/e2e tests, CI/CD, and security hardening (SSRF guards, input fencing, atomic writes).