1
Intake
Spine pulls updates from Notion, WordPress, GitHub, HubSpot, and other connected systems via connectors and webhooks.
2
Load
Raw records are staged in the ingestion layer with source metadata and timestamps.
3
Normalize
Source-specific fields are mapped to the shared domain model: Article, Asset, Campaign, Author, and so on.
4
Schema Mapping
The Schema Agent validates that normalized records conform to the Content domain contract. Misclassified items are flagged.
5
Context Assembly
Spine builds practitioner-specific context by linking related articles, campaigns, topics, and deadlines on demand.
6
Projection
The assembled context is rendered into views: Editorial Calendar, Content Pipeline, Knowledge Base, and others.
7
OODA
Twin and domain agents observe changes, orient them to the domain model, decide on proposals, and surface actions.
8
Governance
Every proposed action is checked against capability authorization, scope, and approval rules.
9
Execution
Approved actions are dispatched to the target connector for source-system writeback.
10
State Update
The source system confirms the change. Spine emits an event, updates views, and preserves the receipt.