Eventcatalog
The discovery and governance layer for event-driven systems. Document your domains, services, events and schemas — for your teams and your AI agents.
Last verified:
What is Eventcatalog?
EventCatalog is a self-hosted, open source discovery and governance layer for event-driven systems. It turns distributed architecture into a searchable, visual, and AI-queryable source of truth. The tool helps architects and engineering teams find what events exist, see how services connect, document domains and schemas, and understand the impact of changes before making them.
Key features include markdown-powered docs-as-code for architecture, automated visualization of service/event/message connections, natural language questioning about your architecture, ownership tracking to know who owns what, and built-in MCP Server support for AI agents. It generates documentation from OpenAPI, AsyncAPI, and schema registries, and produces llms.txt for AI agent discovery.
EventCatalog is designed for software architects, platform engineers, and teams running event-driven architectures at scale. It models architecture using software primitives like domains, services, messages, and schemas rather than generic documentation pages. Companies like Nike, AWS, GOV.UK, M&S, and Eurostar use it to document their systems.
The tool is fully open source, self-hosted, and bootstrapped with no VC funding. Your architecture data never leaves your infrastructure, and you can bring your own LLM models. Over 33,000 catalogs have been created with 2,700+ GitHub stars and 1,200+ community members.
Eventcatalog pricing
Pricing model: Freemium
EventCatalog is open source and self-hosted at no cost. The Free tier is perfect for getting started with architecture modeling at $0. There is also a Pro plan at $12/editor/month with a 14-day free trial for engineering teams designing architectures. The Enterprise plan offers custom solutions for large organizations. The Studio version has tiered pricing: Starter plan at $199/month for small teams building their first governed catalog (includes everything in Community Plan), Scale plan for growing platform teams, and Enterprise for custom solutions. Early Access Special offers discounted pricing for the first year on annual plans.
Eventcatalog pros
- Open source and self-hosted with no vendor lock-in
- Docs-as-code approach using markdown in Git repositories
- Built-in MCP Server for AI agent integration
- Natural language questions about architecture in plain English
- Automatically generates from OpenAPI and AsyncAPI specs
- Visualizes how every service, event, and message connects
- Shows ownership and impact before making architecture changes
- Generates llms.txt for AI agent discovery automatically
- Technology-agnostic, works with any architecture stack
- Supports domain-driven design primitives as first-class resources
- 33,000+ catalogs created showing strong adoption
- 2,700+ GitHub stars from active community
- Bootstrapped with no VC pressure on roadmap
- Enterprise customers include Nike, AWS, GOV.UK
- SDK and API support for automation and CI/CD integration
- Version-controlled documentation with Git
- Shows dependencies and what will be affected by changes
- Works with Cursor, Claude, VS Code, Windsurf
- Bring your own LLM models for privacy
- Schema change detection with automatic docs updates
Eventcatalog cons
- Requires Node.js >= 16.14.0 and Yarn >= 1.5
- Needs server deployment since it uses server-side file access
- Learning curve for teams new to docs-as-code workflows
- Self-hosting requires infrastructure management
- Primarily focused on event-driven architectures
- Beta Editor may have incomplete features
- Manual markdown editing for non-automated content
- No managed cloud hosting option for the open source version
- Starter plan at $199/month may be expensive for small teams
- AsyncAPI great for individual services but discovery still challenging
Frequently asked questions about Eventcatalog
What is EventCatalog?
EventCatalog is a self-hosted, open source discovery and governance layer for event-driven systems. It turns your distributed architecture into a searchable, visual, AI-queryable source of truth. It helps you find what exists, see how it connects, and change it safely as you grow. It is built for architects and AI agents.
Is EventCatalog free?
Yes, EventCatalog is open source and free to self-host. The Free tier is perfect for getting started with architecture modeling at $0. Paid tiers include Pro at $12/editor/month for teams and Enterprise with custom solutions. The open source version has no lock-in and you can self-host it anywhere.
How do I install EventCatalog?
The easiest way is to use the CLI: npx @eventcatalog/create-eventcatalog@latest [name]. This scaffolds a skeleton EventCatalog website. Requirements are Node.js >= 16.14.0 and Yarn >= 1.5. Then run npm run dev to start the development server at localhost:3000.
What architectures does EventCatalog support?
EventCatalog is technology-agnostic and supports any architecture. It models domains, services, messages, schemas, teams, and flows as first-class resources. It works with DDD practices, APIs, brokers, and captures bounded contexts regardless of your stack today or what you adopt next.
How does AI integration work?
EventCatalog has a built-in MCP Server that lets AI agents like Cursor, Claude, VS Code, and Windsurf ask questions about your architecture. It generates llms.txt automatically so any AI agent can discover your architecture. You can bring your own models (OpenAI, Anthropic, or any LLM) and your data stays on your infrastructure.
Can I automate catalog generation?
Yes, EventCatalog syncs with OpenAPI, AsyncAPI, and schema registries to generate catalog resources automatically. You can run generation locally or in CI/CD with one command. The SDK and API support connects internal systems for automation.
What companies use EventCatalog?
Companies like Nike, AWS, GOV.UK, M&S, Eurostar, and many others trust EventCatalog. Enterprise customers exist in healthcare, banking, and transport sectors. Over 33,000 catalogs have been created worldwide with 2,700+ GitHub stars.
How do I add events to my catalog?
Events go in the /events directory. Create a folder with your event name and an index.md file inside (e.g., /events/UserSignedUp/index.md). Fill it with front-matter data (name, version, summary, consumers, producers) and markdown content. Run EventCatalog and navigate to localhost:3000/events/YourEvent.
Is my data secure with EventCatalog?
Yes, EventCatalog is self-hosted and secure by default. Your architecture, queries, and AI context all stay on your infrastructure. Your data never leaves your environment. You own your models and your data with no external hosting required.