Cloud extractor — Confluence pages to ArchiMate Business layer

[victorsoriadrp]

Context

Confluence is where most organizations document Business architecture (capabilities, processes, value streams). A Confluence extractor lets architects pull that semi-structured content into ArchiPulse as Business layer elements without manual re-entry.

Scope

• Authenticate via Confluence Cloud API token
• Discover pages in a configurable space / label filter
• Map page hierarchies to capability trees (configurable mapping)
• Extract structured data from page properties macro
• Preserve back-references (each element carries source: confluence:<pageId>)

Acceptance criteria

• CLI command: archipulse extract confluence --workspace <id> --space <key>
• MCP tool: extract_confluence_pages
• Mapping config (YAML): page labels → ArchiMate element types
• Idempotent re-runs
• Snapshot before import
• Documented (linked to Documentation site — architecture, deployment, API and MCP reference #92)