What problem does it solve? Drawing a correct DDD context map by hand is error-prone: edges go unlabeled, aggregates leak across boundaries, and shared kernels get buried inside one context. This Skill produces or audits a strategic context map where every bounded context is its own frame, every cross-context edge carries a relationship pattern, and the result passes automated lint, scoring, and architecture validation. ## Core Features & Use Cases - Context map generation: Creates one frame per bounded context with aggregates, entities, value objects, the shared kernel straddling its two co-owners, and labeled domain-event arrows, using the Excalidash MCP tool chain. - Relationship labeling: Marks every inter-context edge with exactly one pattern (Partnership, Shared Kernel, Customer/Supplier with U/D, Conformist, Anticorruption Layer, Open Host Service, Published Language). - Reverse engineering and audit: Derives contexts and shared kernels from a codebase via repo analysis, or audits an existing map for unlabeled edges, leaked aggregates, dead events, and missing ACLs, then repairs until the score reaches 95 with zero blockers. - Use Case: Ask it to map an e-commerce domain with Ordering, Payments, Inventory, and Shipping contexts; it draws the frames, wires OrderPlaced and PaymentCaptured events, marks the Ordering-to-Inventory ACL, validates the architecture, and exports the drawing. ## Quick Start Ask the agent to draw a DDD context map for your domain, naming each bounded context, its aggregates, the shared kernel, the domain events, and the relationship pattern on each connection.
