What problem does it solve? Developers receiving a DIVA user story or a myService anomaly ticket often do not know where to start in the Divalto Harmony ERP codebase. This Skill orchestrates a collaborative pre-action analysis that produces a preparation report with similar code examples, reusable DIVA language functions, impact studies, and concrete action points before any code is written. ## Core Features & Use Cases - Four-phase pipeline with 5 human checkpoints: parses the request, queries the Neo4j diva-mcp graph (X.12 advisory snapshot), verifies candidates against the X.13 standard source tree, then builds a facts.json plus a three-layer markdown deliverable. - Hybrid search strategy: prioritizes the Neo4j graph for candidate discovery, then systematically confirms, flags as disappeared, or discovers new findings in the X.13 ERP sources, with graceful degradation to direct mode when the MCP is unavailable. - Selective report content: includes only the relevant content types from a 9-type catalog (code examples, impact study, action sites, business constants, dossier parameters) with traceability in metrics.json. - Use Case: A developer pastes a user story about adding a payment family field to the customer record; the Skill parses it, finds similar DIVA code in the ERP, verifies it in X.13, and delivers a report with concrete files and functions to modify. ## Quick Start Paste your DIVA user story or myService ticket into the conversation and ask Claude to analyze it with the analyzing-diva-request skill to get a pre-action preparation report.
