Let's get you started. This isn't a rigid tutorial—think of it as a friendly walkthrough. You'll feel how BigRack works, then you'll naturally find your own flow.
Want to see your projects and tasks visually? Start the dashboard with this command:
bigrack gui startThe dashboard opens automatically in your browser. Perfect for exploring your repos, projects, and tasks visually.
First things first—let's set up BigRack on your machine. This is a one-time thing:
bigrack initThis sets up the BigRack database and downloads the embedding model (~80MB).
A Repo is where your project's context lives. Think of it as your project's memory. Let's create one:
The easiest way? Just ask your AI. It knows how to use bigrack_create_repo:
Just say:
"Create a new BigRack repo for this project"
Note: Repo creation via CLI is not yet available. Please use the bigrack_create_repo MCP tool. CLI commands for repo management are planned for a future release.
This creates a bigrack.json file and registers your Repo in the local database (~/.bigrack/).
Now let's give your project some memory. Add the rules, patterns, and context that matter:
Just talk to your AI naturally. It'll use bigrack_store_context behind the scenes:
Just say:
"Store a business rule: Stock cannot be negative. Priority: critical"
Note: Context management via CLI is not yet available. Please use the MCP tools in your AI assistant (Claude Desktop, Cursor, etc.) to add business context. CLI commands for context management are planned for a future release.
Projects are where the magic happens. Each one is a focused work unit—a feature, a bugfix, whatever you're building. Let's create one:
Again, just ask your AI. It handles bigrack_create_project for you:
Just say:
"Create a new project called 'Payment Integration' of type feature"
Note: Project creation via CLI is not yet available. Please use the bigrack_create_project MCP tool in your AI assistant. CLI commands for project management are planned for a future release.
The Project automatically inherits business context from the parent Repo.
Got a big feature? Break it down naturally. Your AI uses bigrack_decompose_feature to turn complexity into manageable tasks:
Just say:
"Decompose the feature 'Payment Integration' into atomic tasks with their dependencies"
BigRack reads your context, understands the dependencies, and builds a smart task graph. It just works.
Need to remember something? Just ask. BigRack's semantic search finds what you need:
Your AI automatically uses bigrack_query_context when you ask questions:
Just ask:
"What are the authentication requirements for this project?"
Behind the scenes, BigRack uses vector embeddings to find exactly what you need. It's like having a perfect memory.
Note: Context querying via CLI is not yet available. Please use the bigrack_query_context MCP tool in your AI assistant. CLI commands for context querying are planned for a future release.
Wondering what to work on next? BigRack knows. It tracks dependencies and suggests the right tasks:
Your AI uses bigrack_get_next_step to give you smart recommendations:
Just ask:
"What should I work on next for the Payment Integration project?"
bigrack statusYou can also use bigrack ticket next to get recommended next tasks from the command line.
You've got the basics. Now explore deeper, find your flow, and make BigRack yours: