context introspect

The introspect command connects to a database, reads its schema, and generates .context.yaml files with Bronze-level metadata.

Usage

context introspect --db <connection-url>

Options

Option	Description	Default
--db <url>	Database connection string	$DATABASE_URL
--source <name>	Logical name for this data source	Inferred from DB name
--tables <glob>	Filter tables by glob pattern	* (all tables)
--model-name <name>	Override the model name in output files	Table name
--select	Interactively pick which tables to include	Off (all tables)

Examples

Introspect all tables

context introspect --db postgres://localhost:5432/myapp

Introspect specific tables

context introspect \
  --db postgres://localhost:5432/myapp \
  --tables "users,orders,order_*"

Interactive table selection

context introspect \
  --db postgres://localhost:5432/myapp \
  --select

This presents a checklist of all tables in the database, letting you pick which ones to include. Useful for large databases where you only need a subset of tables.

Name the source

context introspect \
  --db postgres://localhost:5432/myapp \
  --source production

This produces files namespaced under context/production/.

Override model name

context introspect \
  --db postgres://localhost:5432/myapp \
  --tables "user_accounts" \
  --model-name users

Output

For each table, introspect creates a .context.yaml file containing:

• Table name and schema
• Column names, types, and nullability
• Primary keys and unique constraints
• Foreign key relationships
• Indexes

# context/public/users.context.yaml (generated)
model:
  name: users
  source: myapp
  schema: public
  columns:
    - name: id
      type: integer
      primary_key: true
      nullable: false
    - name: email
      type: varchar(255)
      nullable: false
      unique: true
    - name: created_at
      type: timestamp
      nullable: false
  relationships:
    - name: orders
      type: has_many
      foreign_key: user_id
      references: orders

Supported databases

• DuckDB (duckdb://path.duckdb)
• PostgreSQL (postgres://...)
• MySQL (mysql://...)
• SQL Server (mssql://...)
• SQLite (.sqlite, .sqlite3 files)
• Snowflake (snowflake://account/database/schema)
• BigQuery (bigquery://project/dataset)
• ClickHouse (clickhouse://host:port)
• Databricks (config file only)

Each adapter requires its own driver as an optional peer dependency. See Database Support for installation details.

Smart driver detection: If the required driver is not installed, context introspect (and context setup) will detect this and offer to install it automatically.

Next steps

• Enrich the generated files with context enrich
• Validate them with context lint