ADF JSON Example Query Skill

This skill fetches the raw Atlassian Document Format (ADF) JSON data from a Confluence page. Use this when you need to understand the actual ADF structure of a real Confluence page, inspect how specific elements are represented in ADF, or debug ADF parsing issues.

Purpose

When working with ADF data, you often need to see real-world examples of how Confluence represents content. This skill:

• Fetches the complete ADF JSON body from any accessible Confluence page
• Displays the raw JSON structure with proper formatting
• Caches results locally to avoid repeated API calls

Prerequisites

The Confluence API credentials must be configured. See atlas_doc_parser/tests/data/client.py for authentication setup.

Command

Fetch ADF JSON from a Confluence page URL:

.venv/bin/python .claude/skills/adf-json-example/scripts/adf_json_example_cli.py get_example "<confluence_page_url>"

Parameters

• confluence_page_url: Full URL to a Confluence page (e.g., https://your-domain.atlassian.net/wiki/spaces/SPACE/pages/123456/Page+Title)

Example

.venv/bin/python .claude/skills/adf-json-example/scripts/adf_json_example_cli.py get_example "https://sanhehu.atlassian.net/wiki/spaces/GitHubMacHuGWU/pages/654049306/Mark+-+strong"

This outputs the full ADF JSON structure of the page, showing how nodes and marks are organized.

Output

The command prints formatted JSON to stdout, containing the full ADF document structure with:

• version: ADF format version (typically 1)
• type: Document type (typically "doc")
• content: Array of top-level nodes (paragraphs, headings, tables, etc.)

Each node contains its type, attributes, and nested content/marks.