Detect phantom and unused dependencies across multiple languages
One CLI finds phantom and unused deps across five languages when per-language tools already exist.
๐ Detect architectures, platforms, shells, terminals, CI systems and agents, grouped by family
Zero-dependency Python library replacing removed stdlib functions with comprehensive platform detection.
Python developers and DevOps engineers
Distro ยท platform
It detects six traits (CPU architecture, OS/distribution, shell, terminal, CI, agents), which are grouped into families (BSD, LINUX, UNIX, ...) for convenience.
You can try the library without installing anything:
$ uvx --with extra-platforms python >>> from extra_platforms import current_platform, BSD, is_linux >>> current_platform() Platform(id='macos', name='macOS') >>> current_platform() in BSD True >>> is_linux() False
It also ships Pytest decorators (@skip_linux, @unless_macos) for platform-conditional tests.The library has zero dependencies and is Apache-2.0 licensed.
I'm interested in collecting edge cases, which you can send me by performing an auto-detection with:
$ uvx extra-platforms
One CLI finds phantom and unused deps across five languages when per-language tools already exist.
Yet another LLM ops layer when LangSmith, Helicone, and Braintrust already exist.
Single-purpose CLI for a solved problem; remove.bg has better use cases.
Compressed JSON bundles fit tight context windows better than pasting files.
Tech stack detector powered by RapidAPI; Wappalyzer and StackShare do this already.
Phonetic embeddings catch ASR-mangled names across cultures before LLM sees them.