1. Create the project

You are not building a pipeline, so you do not need a platform-specific template. An empty project is all the context layer needs:

bruin init empty ai-analyst
cd bruin

That creates:

bruin/                  # project root, initialised as a git repository
└── ai-analyst/         # the pipeline folder
    ├── pipeline.yml    # a name, with everything else commented out
    └── assets/         # your context layer lives here

Two surprises worth naming. init wraps your pipeline in a parent folder called bruin - that is the project root, which is why you cd bruin rather than cd ai-analyst. And it runs git init there, so the commits later in this guide work without any setup. Pass --in-place if you would rather have ai-analyst/ created directly in the current directory.

.bruin.yml, the file that holds your credentials, does not exist yet. It is created the first time you add a connection, and it is gitignored.

Two paths matter for the rest of this guide, and mixing them up is the most common mistake:

• .bruin.yml sits at the project root, next to the ai-analyst/ folder.
• ai-analyst/ is the pipeline folder, the one containing pipeline.yml. This is the path you pass to import and enhance.

assets/ ships with an empty placeholder file so the folder is not lost in git. Delete it now, so the import in step 3 is the only thing in there:

rm ai-analyst/assets/placeholder