Inspecting Schemas

The pmb schema command inspects SQLAlchemy model definitions to display schema, table, and column information. No database connection required -- all data is read from the model definitions.

Usage

# List all schemas
pixi run pmb schema

# List tables in a schema
pixi run pmb schema pdbj

# Show columns in a table
pixi run pmb schema pdbj.brief_summary

# Show single column detail
pixi run pmb schema pdbj.brief_summary.pdbid

Search

Search column names and comments across all schemas:

# Search by column name
pixi run pmb schema -s resolution

# Search by comment text
pixi run pmb schema -s "deposition date"

JSON Output

The --json flag outputs machine-readable JSON, useful for AI tools and scripts:

# All schemas as JSON
pixi run pmb schema --json

# Full schema with all tables and columns in one response
pixi run pmb schema pdbj --json

# Table columns as JSON
pixi run pmb schema pdbj.brief_summary --json

# Search results as JSON
pixi run pmb schema -s resolution --json

tip

pmb schema pdbj --json returns all tables and columns in a single JSON response. This is large but ideal for AI agents that need full schema context.

Options

Option	Short	Description
--search	-s	Search column names and comments across all schemas
--json		Output in JSON format (machine-readable)
--config	-c	Config file path (default: config.yml)

Examples

# Find all columns related to resolution
pixi run pmb schema -s resolution

# Check what columns are in the entity table
pixi run pmb schema pdbj.entity

# Get full cc schema for AI processing
pixi run pmb schema cc --json

# Look up a specific column's type and comment
pixi run pmb schema pdbj.brief_summary.pdbid