> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nodaldata.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Build your context

> Run the interview to generate your first ACF domain — free and local.

<Info>Starter page — expand with screenshots and a worked domain walkthrough.</Info>

The `context-interview` skill builds your context layer *with* you, one domain at a time, and writes
it to a reviewable git repo.

## Prerequisites

* A [warehouse connected over MCP](/open-source/self-service-mcp) (read-only).
* *(Recommended)* your dbt repo cloned locally, so the interview reads real `models/` and
  `schema.yml` as a draft to confirm.

## Run it

```bash theme={null}
# Clone the whole repo — the skill reads SPEC.md, template/, schemas/, and scripts/.
git clone https://github.com/nodal-data/nodal-context.git
cd nodal-context

# Open your agent (Claude Code / Codex / Cursor) here and say:
#   "Build my analytics context."
# The context-interview skill takes over and writes a reviewable ../analytics-context/
# sibling repo (git-initialized), then offers to push it to GitHub.
```

## What you get

A `../analytics-context/` repo — your deliverable — containing ACF domains the analyst confirmed.
The cloned tool repo stays read-only; keep it to add more domains later.

## Next

* [See the accuracy delta](/open-source/evaluation) your context produced.
* [Share it with your team](/mcp/share-with-your-team) over the hosted MCP.