Quickstart
This walkthrough uses only the public synthetic fixture.
Generate and run the fixture
brushline fixture --out tmp/corridor
brushline prototype --fixture tmp/corridor --out tmp/prototype --years 5
The first command writes a deterministic two-corridor fixture. The second compiles it into a
validated dataset and writes a five-year prototype run under tmp/prototype.
Inspect the result:
find tmp/prototype -maxdepth 3 -type f | sort
sed -n '1,80p' tmp/prototype/run/reports/summary.md
The run directory contains the compiled dataset, the simulated run, the schedule, and a summary report. The exact filenames are defined by the artifact contracts in Artifact contracts.
Build the static demo
brushline demo build --out build/demo
python -m http.server 8000 --directory build/demo
Open http://localhost:8000/. The static demo is read-only and uses the checked synthetic example.
Run the studio
python -m uvicorn brushline.web.studio:app --host 127.0.0.1 --port 8801
Open http://127.0.0.1:8801/. With no private artifacts present, only the synthetic and observed
public instances are available. The network instance appears only when its ignored private artifact
set exists.
Private network prototype
The BC network prototype is a separate, private workflow. It requires the cached OSM, VRI, DEM, and network artifacts described in Build the network prototype. Do not run the network builder on a clean public checkout and do not publish its output.
Next steps
Your first study explains what the generated artifacts mean.
Concepts explains the model and planning stack.
Limitations and boundaries states what the prototype does not claim.