Event-Driven Architecture Designer

Part of the worlds-biggest-software-project initiative.

An AI-native, open-source visual designer for event-driven architectures, with schema registry integration and consumer tracking across heterogeneous brokers.

The Event-Driven Architecture Designer is a design-time and governance tool for engineering teams building microservices and AI agent systems on event-driven backbones. It unifies AsyncAPI authoring, schema registry integration, topology visualisation, and consumer impact analysis in one developer-friendly catalog, addressing the governance pain that emerges as event catalogs grow to thousands of event types.

Why Event-Driven Architecture Designer?

Existing open-source documentation tools such as EventCatalog cover catalog and visualisation but lack native real-time topology discovery and direct schema evolution impact analysis.
Confluent Schema Registry is the industry standard for schema management but offers no documentation, ownership metadata, visual topology, or consumer impact analysis, and is Kafka-centric.
AWS EventBridge Schema Registry is AWS-only, supports only OpenAPI 3 / JSONSchema Draft4, and enforces no compatibility checks.
Solace PubSub+ Event Portal is the most complete commercial designer but is proprietary, expensive, primarily Solace-focused, and lacks AI-native features.
General diagramming tools (Miro, Lucidchart) are widely used as informal stand-ins but have no schema awareness or live topology data.
No incumbent works equally well across Kafka, RabbitMQ, NATS, and Azure Service Bus, leaving cross-broker governance and orphaned-event detection underserved.

Key Features

Catalog and Specification Authoring

AsyncAPI 3.x specification authoring with live validation and rendered documentation preview
Multi-service event catalog covering domains, services, events, schemas, and their relationships, stored in Git
Markdown-and-Git workflow so all content lives in source control and is developer-friendly
Data Products as a first-class resource type linking event streams to analytics-ready tables

Schema Management and Governance

Import from Confluent Schema Registry, AWS Glue Schema Registry, and Apicurio Registry
Backward, forward, and full schema compatibility checking at design time
Field-level schema search and type-conflict detection across all message schemas
Schema change consumer impact analysis flagging all consumers of a schema before a breaking change ships
Lifecycle management with semantic versioning and governed promotion stages

Topology Visualisation and Runtime Sync

Visual topology viewer showing producer to event to consumer dependency graphs
Runtime topology discovery by connecting to live Kafka clusters and comparing against the designed catalog
Architecture change detection comparing catalog state across git branches and against live runtime state
Orphaned event detection identifying events with no active consumer or producer

AI-Native Assistance

AI assistant for querying the catalog in natural language, exposed via an MCP server
AsyncAPI spec auto-generation by analysing existing Kafka topic and consumer group metadata
Natural-language-to-schema design that accepts a business process description and proposes an AsyncAPI spec
Event ownership recommendation derived from git commit history and CODEOWNERS analysis

Cross-Broker and Integration Support

Kafka, RabbitMQ, NATS JetStream, and Azure Service Bus support beyond the Kafka-centric incumbents
CloudEvents envelope support for vendor-neutral event interoperability
REST API and TypeScript SDK for programmatic catalog access and custom plugins
Contract testing integration compatible with Microcks for CI/CD pipeline validation

AI-Native Advantage

Where incumbents stop at static catalogs and registry CRUD, this project layers AI capabilities directly on the architecture graph. Natural-language event design lets architects describe a business process and receive a proposed event schema, topic structure, and consumer dependency graph. Schema evolution impact analysis automatically identifies downstream consumers, assesses compatibility, and ranks migration risk before a change is deployed. Anomaly detection on consumer lag, event rates, and schema validation failures catches architectural drift before it causes outages, and AI-powered ownership recommendations surface orphaned events that have lost their producing or consuming team.

Tech Stack & Deployment

The project centres on AsyncAPI 3.x as the canonical design format and CloudEvents as the event envelope. Schema management integrates with Confluent Schema Registry, AWS Glue, and Apicurio across Avro, Protobuf, and JSON Schema. Deployment is expected to support self-hosted (Git-backed, suitable for regulated and air-gapped environments) and cloud modes, with an MCP server exposing the catalog to Claude, GitHub Copilot, and other AI tools, and a TypeScript SDK plus REST API for plugin development.

Market Context

The global event-driven architecture and messaging middleware market is estimated to exceed $10 billion in 2026, driven by microservices, real-time data, and AI agent architectures (RisingWave, 2026). Purpose-built EDA design tooling spans free open source (EventCatalog) to custom enterprise pricing (Solace PubSub+ Event Portal), with general diagramming tools at $8–15/user/month often used as low-cost stand-ins. Primary buyers are principal engineers and architects, platform teams responsible for event-broker governance, integration architects at large enterprises, and data engineers building real-time streaming pipelines.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.