- Home
- Skills
- Product Management
- Event Storming
Event Storming
This is Alberto Brandolini's Event Storming, adapted for an AI-assisted conversation.
$5
Works with the AI Agents you already use
Event Storming
Example session with this skill installed
Run event storming on J1 Fleet Daily Monitoring.
- Read your context and instructions
- Compiled the event storming
The skill loads J1, features F1–F3, US1–US6 and personas P1/P2. It extracts PositionUpdated, GeofenceViolated and AlertCreated as events, UpdatePosition and DefineGeofence as commands, VehicleTrackingView as a read model and TelematicsAPI as an external system. It proposes one BC (Fleet Monitoring), then asks whether a geofence can change independently of a vehicle. After confirming the Vehicle and Geofence aggregates, their invariants and GeofenceCheckPolicy, it writes context-map.md, event-storming-fleet-daily-monitoring.md, fleet-monitoring/README.md (PlantUML) and aggregate-vehicle.md / aggregate-geofence.md.
Connects securely to your tools. The creator never sees your data.
About this skill
Event Storming
The domain-modeling layer of the Harness Skills product-management pipeline: Journey + Features + User Stories → Event Storming → DDD Domain Model.
This is Alberto Brandolini's Event Storming, adapted for an AI-assisted conversation. The skill splits the work by how much user input each element needs:
- It extracts the low-input elements itself: events, commands, read models and external systems, all from your acceptance criteria and journey touchpoints.
- It elicits the high-input ones: aggregate boundaries, invariants and policy rules, one question at a time.
What it does
- Journey-first scoping: pick a journey; it loads the linked features, user stories, personas, capability and vision constraints.
- User Story completeness gate: it blocks or warns when features have no stories with Given/When/Then acceptance criteria.
- Mechanical extraction:
- "Then X is created" →
XCreated - "I want to…" → commands
- "So that I can see…" → read models
- Read-only actions are suppressed as events.
- "Then X is created" →
- Bounded Context framing first: BCs come from capability boundaries plus a language-divergence check, and a context map is built from the journey's system touchpoints.
- Adaptive path: a fast path for simple domains (1-2 aggregates); interactive elicitation for complex ones (aggregate boundaries, invariants, policies, ubiquitous language).
- Formalization: every concept maps to DDD annotations, including:
@AggregateRoot@ValueObject(ID/VALUE/CRITERIA/ENUM)@DomainCommand/@DomainEvent@DomainService(POLICY/SPECIFICATION/DOMAIN)@Repository,@Factory,@Provider,@ReadModel
- Derived fields and behaviors: aggregate fields come from VOs, command parameters and event payloads. Methods come from commands, guarded by invariants.
- Diagrams: a PlantUML domain overview per BC, PlantUML per aggregate, a Mermaid context map and a Mermaid event flow.
- Bidirectional traceability: every command, event and invariant has a Source (
US{n}/J{n}.Phase{m}/F{n}). A reverse Traceability Index lets you look up concepts by Journey, Feature or User Story. - Quality gates: event/command/actor/policy coverage, no orphans, one ID per aggregate, at least one invariant, a size limit and CQRS separation.
Output
product-management/domain/
context-map.md # BCs, Mermaid map, traceability index
event-storming-{journey-slug}.md # discovery audit trail
{bc-slug}/README.md # UL, cross-aggregate services, PlantUML overview
{bc-slug}/aggregate-{slug}.md # one file per aggregate
Package
SKILL.md, README.md (methodology and design rationale), 4 references (DDD annotation catalog, enrichment protocol, interaction protocol, state panels) and 4 output templates.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Product Management.
It requires upstream artifacts (at least one journey plus features, ideally user stories with acceptance criteria) and refuses to start without them. Run product-framer → journey-designer → user-story first. Some templates and the journey-layer labels are in Simplified Chinese; the conversation follows the user's language. It produces design documents, not code. The annotation vocabulary follows the MIT ddd-concepts Java library, but the output is language-agnostic markdown.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
- Software
- Upstream journeysfeatures
Permissions
File Scopes
No network access. Never modifies upstream artifacts.