01
Fast feedback loops
Projects, tests, formatting, linting, profiling, and benchmarks are already part of the repo workflow.
Arden combines LLVM-backed native output, static safety checks, and an integrated command-line workflow so teams can move from experiments to multi-file projects without swapping mental models.
Repository-first workflow
Fast path from zero to project mode
Command flow
$ arden new radar
$ cd radar
$ arden check
$ arden test
$ arden runWhat this unlocks
01
Projects, tests, formatting, linting, profiling, and benchmarks are already part of the repo workflow.
02
Arden targets LLVM directly and keeps the surface biased toward practical systems work over unnecessary abstraction.
03
Ownership and static checks are there to prevent damage early, without turning every function into ceremony.
Core capabilities
The repo is strongest when the language, examples, docs, and tooling reinforce each other instead of looking like separate side projects.
Browse the repositoryOwnership, borrowing, mutability, and semantic validation push failures to compile time before they leak into runtime debugging.
Build, run, check, test, fmt, lint, fix, bench, profile, bindgen, parse, lex, and LSP support already sit in one workflow surface.
Multi-file builds, explicit source graphs, and reusable cache state make the repo feel like an actual language toolchain, not a parser demo.
Installation
Grab the matching portable bundle, unpack it, run the included launcher, and move straight into compiling. The repo docs stay there if you want the deeper source-build path.
Documentation
Every markdown page under `docs/` is copied into the web build and indexed into documentation navigation, so the site stays aligned with the repo instead of drifting behind it.
Entry point
Arden Overview
Entry point
Projects