datajunction

DataJunction semantic layer expert. Use proactively for any DataJunction or DJ work — querying metrics, exploring nodes and dimensions, building SQL, understanding lineage, authoring metrics, and semantic layer design.

Activate this skill when authoring DataJunction (DJ) nodes via the REST API directly (curl, HTTP clients) — typically for exploration, ad-hoc prototyping, or namespaces that aren't repo-backed. For modeling decisions and the decomposition workflow, invoke `datajunction-semantic-model`. For repo-backed YAML authoring (the production path), invoke `datajunction-repo`. For concepts, invoke `datajunction`. Keywords: - DJ API, REST API, curl - POST nodes/metric, POST nodes/dimension - create metric, create dimension, create cube - API approach, direct API changes - prototyping, exploration

Activate this skill for querying DataJunction (DJ) — finding nodes, generating SQL, fetching metric data, exploring lineage, visualizing results — via the DJ UI, MCP tools, or REST/GraphQL APIs. For core DJ concepts and vocabulary, invoke `datajunction`. For modeling decisions (what shape something should take), invoke `datajunction-semantic-model`. For authoring nodes, invoke `datajunction-repo` (YAML) or `datajunction-api` (REST). Keywords: - query metric, query metrics - generate SQL, build metric SQL - get metric data, fetch metric - available dimensions, common dimensions - search_nodes, get_node_details, get_node_lineage - get_common, build_metric_sql, get_metric_data - visualize metrics - MCP tools, DJ API, GraphQL - DJ UI, web UI, browse

Activate this skill when authoring DataJunction (DJ) nodes via YAML files in a git repository — the repo-backed workflow. Covers YAML schemas per node type, branch-based development, temporal partitions on cubes, and the full PR-driven deployment flow. For modeling decisions (how to structure metrics, decomposition workflow), invoke `datajunction-semantic-model`. For direct API authoring, invoke `datajunction-api`. For concepts, invoke `datajunction`. Keywords: - YAML nodes, YAML definitions - repo-backed namespace, repo-backed workflow - git workflow, branch development, feature branch - cube YAML, metric YAML, dimension YAML, transform YAML - create metric, create dimension, create cube, build cube - temporal partition, partition pushdown - pre-commit, push.sh

Activate this skill for DataJunction (DJ) semantic modeling decisions — choosing the right node shape (fact, dimension, transform, metric, cube), turning a draft SQL query into well-designed nodes, and the cross-cutting conventions (ownership, naming, namespace organization). Format-agnostic modeling guidance; for YAML schemas and the repo-backed authoring flow, invoke `datajunction-repo`; for direct API examples, invoke `datajunction-api`. For DJ concepts (node types, dim links), invoke `datajunction`. For querying existing metrics, invoke `datajunction-query`. Keywords: - semantic modeling - decompose query, model query, query to nodes - create metric, create dimension, create node, create cube, build cube - ratio metric, derived metric, base metric - composable metrics - metric query constraints - node ownership, metric ownership - metric naming, namespace organization - grain, fact vs dimension - dimension link, not JOIN

Activate this skill whenever working with DataJunction (DJ) semantic layer. Core concepts and shared vocabulary used by every DJ workflow. For querying metrics, invoke `datajunction-query`. For modeling decisions (what shape something should take), invoke `datajunction-semantic-model`. For authoring nodes, invoke `datajunction-repo` (YAML in a git repo) or `datajunction-api` (REST API for exploration / prototyping). Keywords: - DataJunction, DJ - semantic layer - dimension link, dimension links - star schema - node types - source, transform, dimension, metric, cube - metric, metrics - mode, status, valid, invalid, draft, published - namespace