Tutorial
Reproduce the research workspace
This first-run tutorial verifies the pinned tools, shows what the workspace actually registers, and builds the current research corpus. It is deliberately about learning the workflow, not about installing Pinega Engine.
Prerequisites
Work from a clean checkout of likern/research. Nushell and Typst versions are pinned by repository files; do not substitute assumed system versions when a reproducibility failure is being diagnosed.
1. Verify the workspace
Run the repository doctor before building anything. A successful result establishes that the expected toolchain is available.
^nu research.nu doctor
2. Inspect the registered documents
Ask the workspace for its own document inventory rather than guessing a path or artefact name.
^nu research.nu list documents
The registry output is the authoritative list for the current checkout. This protects the workflow from stale documentation that names a document which has moved or been superseded.
3. Build the registered corpus
^nu research.nu build all
A successful build demonstrates that the checked-in research sources and pinned toolchain can reproduce their registered outputs. It does not validate every research claim; MODEL, VERIFY, tests, and source evidence remain separate gates.
4. Continue from the evidence, not from filenames
Use the research catalogue to choose a subject, then follow its source and artefact links. For repository boundaries use Repository layout; for review rules use Review and release gates.