ADF JSON Schema Query Skill

This skill provides tools to query the Atlassian Document Format (ADF) JSON schema, which defines the structure of all node and mark types used in Confluence pages and Jira issue fields.

Purpose

When implementing ADF nodes or marks as Python dataclasses, you need to understand:

• What properties each type has
• Which properties are required vs optional
• The allowed values for each property

This skill helps you query the schema to get this information.

Commands

List All Definitions

Lists all available node and mark definitions in the ADF schema:

.venv/bin/python ./.claude/skills/adf-format-json-schema/scripts/adf_schema_cli.py list_def

Output format: name = <definition_name>, properties = [<property_list>]

Definitions follow naming conventions:

• *_node - Node types (e.g., paragraph_node, heading_node)
• *_mark - Mark types (e.g., strong_mark, em_mark, link_mark)

Get Definition Details

Gets the full JSON schema for a specific definition:

.venv/bin/python ./.claude/skills/adf-format-json-schema/scripts/adf_schema_cli.py get_def <definition_name>

Example:

.venv/bin/python ./.claude/skills/adf-format-json-schema/scripts/adf_schema_cli.py get_def strong_mark

Workflow

When implementing an ADF node or mark:

1. List definitions to find the exact definition name
2. Get definition details to understand the schema structure
3. Implement the dataclass based on the schema properties and requirements