snowflake-architecture-diagram-from-lineage

From-lineage mode

Generates a diagram of a real, deployed data flow by walking cortex lineage from a target object. Works in any Snowflake account because it depends on lineage metadata only — not on SnowGram-specific deployed objects.

Use cases

• "Show me everything that feeds my dashboard table"
• "Diagram the upstream of MY_DB.MART.SALES_DAILY"
• "What does my fact table depend on, three hops up?"
• "Generate documentation for our actual production pipeline"

Prerequisites

• Active cortex connection
• Current role has USAGE on the target object's database/schema
• Target object exists (table, view, materialized view, dynamic table, or stream)

Path resolution (do this once)

This sub-skill uses $SKILL_DIR/... paths. Resolve them by sourcing the canonical helper. The absolute path to that helper is <SKILL_BASE_DIR>/assets/scripts/skill_paths.sh, where <SKILL_BASE_DIR> is the path the skill loader prints as Base directory for this skill: when this skill is loaded (or the path returned by cortex skill list for snowflake-architecture-diagram).

source "<SKILL_BASE_DIR>/assets/scripts/skill_paths.sh"
# Now: $SKILL_DIR, $TEMPLATES_DIR, $COMPOSER_DIR, $VIEWER_DIR, $SCRIPTS_DIR,
#      $STATE_FILE are all set. See the parent SKILL.md for the full table.

For Python helpers, import assets/scripts/skill_paths.py analogously. The parent SKILL.md shows the importlib pattern.

Workflow

Step 1: Identify the target object

The user should provide a fully-qualified name (DATABASE.SCHEMA.NAME). If they provide a partial name, use cortex search object "<name>" to disambiguate:

cortex search object "<partial name>" --types table,view,dynamic_table

⚠️ STOP: confirm the chosen FQN with the user.

Step 2: Walk the lineage tree

cortex lineage <FQN> --direction upstream --distance 5 --tree

Default direction is upstream (what feeds the object). For reverse exploration ("what consumes this?"), use --direction downstream. For both, run two passes.