setup
Interactive wizard that walks you through the full pipeline. Read more
The ContextKit CLI (context) provides 16 commands organized into three groups: Core Workflow, Exploration, and Serving.
These commands form the main pipeline for creating and maintaining context files.
setup
Interactive wizard that walks you through the full pipeline. Read more
new
Scaffold a new data product inside your project. Read more
introspect
Scan a database and scaffold Bronze-level metadata. Read more
enrich
Suggest or apply Silver/Gold enrichments. Read more
lint
Validate context files against 40 rules. Read more
fix
Auto-fix lint violations where possible. Read more
build
Compile context files into a manifest JSON. Read more
tier
Show an OSI tier scorecard for your context. Read more
Commands for understanding, validating, and verifying context.
explain
Look up any model, term, or owner by name. Read more
rules
List all lint rules with tier and fixability. Read more
validate-osi
Check a file against the OSI specification. Read more
verify
Validate metadata accuracy against a live database. Read more
Commands for making context available to AI tools and humans.
serve
Start an MCP server (stdio or HTTP). Read more
site
Generate static HTML documentation. Read more
dev
Watch mode — re-lint on file changes. Read more
blueprint
Export AI Blueprints — portable Gold-tier data product specs. Read more
init
Scaffold a new ContextKit project structure. Read more
# Full guided setup (recommended for first-time users)context setup
# Or run individual stepscontext introspect --db postgres://localhost/mydbcontext enrich --target silver --applycontext lintcontext buildcontext serve --http --port 3000