Getting started
Install a prebuilt GraphArch release, keep a small publication configuration in your project, and run the same command locally and in CI.
Standalone binaries are available from GitHub Releases. Package-manager channels remain subject to their separate publication status below.
Install a prebuilt binary
Download the archive for your platform from
GitHub Releases. Each archive has an
adjacent SHA-256 checksum file. Verify the download, extract it, and put
grapharch (or grapharch.exe on Windows) in a directory on your PATH.
The release platforms are:
| System | CPU | Requirement |
|---|---|---|
| Linux | x64 or ARM64 | glibc 2.28 or newer |
| macOS | Intel or Apple silicon | macOS 11 or newer |
| Windows | x64 or ARM64 | Native Windows executable |
Musl-based Linux distributions and 32-bit systems are not included in this release matrix. Select the archive for the system where GraphArch will run.
Other installation channels
These channels install the same CLI. They are pending publication; choose the package manager already used by your project once its package is available.
| Channel | Install or run |
|---|---|
| Homebrew | brew install legra-ai/tap/grapharch |
| Python with uv | uv tool install grapharch, or uvx grapharch --help for an isolated run |
| Python with pip | python -m pip install grapharch |
| npm project dependency | npm install --save-dev grapharch, then npx grapharch --help |
| cargo-binstall | cargo binstall grapharch |
For a repeatable CI build, pin an exact released version instead of tracking the latest package. A missing binary or unsupported platform is an installation error; these installers do not silently compile GraphArch from source.
Check the installation
grapharch --version
grapharch publication --help
If the command is not found, check that its installation directory is on your
PATH. In an npm project, use npx grapharch instead of grapharch.
Publish your project
Choose the guide for the publication you want:
| Publication | Guide |
|---|---|
| Related ontologies with linked vocabulary chapters | Ontology families |
| A standards-style specification and PDF | Specification renderer |
| A documentation website of guides and topics | Self-hosted documentation |
| Separate specification, companion website, and PDF | Publication families |
For an ontology family, define your project and target in grapharch.ttl, then
read the existing repository checkout directly:
grapharch publication build-family \
--local-repository . \
--definitions grapharch.ttl \
--project urn:example:project \
--target urn:example:specification \
--output-root output/docs
Replace the example identifiers with those in your configuration. The ontology-family guide explains membership, namespace hints, remote repositories, and the generated files. Source files stay in your project; generated files go under the chosen output directory.
Use the same pinned GraphArch version and publication command in CI. Publish the generated files to GitHub Pages or another static host. Browse the showcase gallery for working output before choosing your publication style.
Build from source
Build from source when contributing or testing unreleased changes. Normal installation uses the prebuilt channels above.
Install the development prerequisites, then run:
git clone https://github.com/EKGF/grapharch.git
cd grapharch
cargo build --release --locked
./target/release/grapharch --help
On Windows, the executable is target\release\grapharch.exe. To generate this
repository's documentation with the built executable:
./target/release/grapharch publication build-site \
--source docs \
--definitions docs/grapharch.ttl \
--project urn:grapharch:docs \
--target urn:grapharch:docs:website \
--output-root output/docs
Open output/docs/site/index.html, or serve that directory with your preferred
local web server. The examples guide contains contributor commands
for running individual fixtures and their tests.
For separate collection, planning, rendering, and inspection commands, see the publication pipeline guide. Check the status page before relying on a planned capability.