Knuckles-Team Confluence Knowledge Ingester Workflow
CONCEPT:KG-2.12
Fetches a Confluence wiki page, saves it locally as a standard markdown file, and pre-stages the artifact for seamless Knowledge Graph ingestion.
Steps
Step 0: Atlassian Agent
Agent: intake-agent
Tools: graph_query, nc_files
Retrieve a Confluence page's HTML or wiki content by ID or space key and title using the atlassian_confluence_page tool.
Expected: confluence_page_data
Step 1: User Interaction
Agent: processor-agent
Tools: graph_analyze, document_tools
Present the fetched Confluence page content summary. Prompt the user for a preferred saving path in the workspace for markdown conversion and future ingestion.
Expected: save_path, file_metadata
Step 2: Graph Os
Agent: validator-agent
Tools: graph_query
Write the formatted page content to save_path and call mcp_graph-os_graph_ingest tool with the target path to natively ingest the Confluence document into the Knowledge Graph.
Expected: ingestion_job_status
Step 3: KG Persistence [depends_on: graph-os]
Agent: validator-agent
Tools: graph_write
Persist workflow results as nodes and edges in the Knowledge Graph. Create appropriate typed nodes with metadata and link to existing domain entities.
Output
- Confluence Knowledge Ingester results persisted in KG
- Structured report (MD/PDF)
- Audit trail with timestamps and agent attributions
Human Oversight Required
✅ Critical decisions require human review and approval.
Execution
Run this workflow as a dependency-ordered DAG. Steps with no unmet depends_on run in parallel; dependents run after their prerequisites complete.
- Run first (in parallel): Step 0 — Atlassian Agent; Step 1 — User Interaction; Step 2 — Graph Os
- After level 0: Step 3 — KG Persistence
Execution: If graph-os is reachable, offload the whole DAG via graph_orchestrate action=execute_workflow (or the kg-delegate skill) for true parallel/swarm execution. Otherwise execute the steps natively in dependency order: run steps with no unmet depends_on in parallel, then their dependents.