Testing phaseThis environment is in pre-launch testing — all data will be reset at launch.
Mockomat
EN English DE Deutsch FR Français ES Español IT Italiano TR Türkçe
v2 · Visual data modelling, live in your browser

Visual data modelling, live API.
Real records, served the moment you save.

Sketch tables and relations with your team on a shared canvas. The model becomes a live GraphQL API backed by real records — no fixtures, no hand-edited JSON.

01 Edit attribute
Identifier: id
displayName
handle
avatarUrl
bio
role
verified
Add attribute
displayName
displayName
Use camelCase (e.g. firstName, createdAt)
STRING
Try it → rename this attribute or remap it to a different mock-dataset field. The JSON and avatars on the right respond live.
02 query users { … } 
{
  "data": {
    "users": [
      {
        "id": "7f2a-…",
        "displayName": "Lena Hartmann",
        "handle": "lena.h",
        "role": "Designer",
        "verified": true
      },
      {
        "id": "a1c4-…",
        "displayName": "Tomás Iglesias",
        "handle": "tomas",
        "role": "Engineer",
        "verified": true
      },
      {
        "id": "b903-…",
        "displayName": "Yuki Tanabe",
        "handle": "yuki",
        "role": "Product",
        "verified": false
      }
    ]
  }
}
03 Rendered in your UI
LH
Lena Hartmann
@lena.h Designer
TI
Tomás Iglesias
@tomas Engineer
YT
Yuki Tanabe
@yuki Product
One canvas · one model · live GraphQL on real records.
No signup · 24h sessions
01 The flow

Three pillars. One canvas.

Visual modelling, real-time collaboration, and a live API on real records — all in one place.

01 Model

Sketch tables and relations on a shared canvas. Live cursors, threaded comments, full history.

02 Map

Wire each field to a real-records source or a fixed value — your choice, per attribute.

03 Serve

A live GraphQL endpoint comes up the moment you save. Backed by real records, queryable from your app.

A new way to model

Database modeling,
finally collaborative.

The first place your team designs schemas the way they design products — together, on a live canvas. Not Slack threads about a screenshot of a spec doc.

  • Live cursors & presence See who's where on the canvas
  • Real-time co-editing Drag, rename, connect — together
  • Comments anchored to attributes Discuss the model, not screenshots
  • Resolve threads inline Decisions live with the schema
Modelling Board · acme · main
L T Y A
4 editing now
User
*id
*displayNameString
*handleString
avatarUrlString
Lena · editing
TeamMember
*id
*teamId
*userId
*roleEnum
joinedAtDT
#1 System Admin −
Lena Hartmann · today, 11:54 LH
Are the values for this enum already reviewed?
Add comment
02 Realistic, on purpose

Data that looks like real users, not Lorem ipsum.

The field catalog describes what the shared pool holds: names, emails, addresses, prices, vehicle makes. Map an attribute to a field and the mock API serves its values from the imported datasets — it invents nothing at request time.

Person
Contact
Location
E-commerce
Vehicle
Real estate
Field catalog
fields of the shared pool
Person first_name → “Lena”
Contact email → “lena.hartmann@northvale.io”
Location address → “42 Rue de Bercy”
Person job_title → “Engineering Manager”
Sample response
200 OK · 6ms
{
“firstName” : "Lena" ,
“email” : "lena.hartmann@northvale.io" ,
“address” : "42 Rue de Bercy" ,
“avatarUrl” : "https://cdn.mock/avatars/lena.png" ,
“jobTitle” : "Engineering Manager"
}
03 The pool, then the pick

One pool, column after column.
Pick the ones that fit your model.

Mockomat's shared pool holds imported datasets: open data where a vetted source exists, generated sample data for the blueprints. You don't seed data; you map it. Point an attribute at a column and the runtime does the rest.

Shared pool One pool for every project · read-only
Composed Coherent rows from pool values · never invented
Constant Pin a fixed value per attribute
mockomat.pool / shared · read-only · shared by every project
commerce person internet address business finance
product_name brand price_eur sku category_main description image_url stock_quantity rating_avg first_name last_name full_name email username avatar_url phone birthdate street city country_iso postal_code lat lng company job_title department salary_band tax_id iban currency transaction_id amount created_at updated_at expires_at tag locale
↓ rows of the imported datasets ↓
pool.full_name
pool.email
pool.avatar_url
map
map
map
UserProfile.displayName STRING
UserProfile.email STRING
UserProfile.avatarUrl STRING
The pool is shared. The picks are yours. The shape is your model's. Read the docs
04 One model, two surfaces

A live API for your code.
A rendered preview for everyone else.

The same model serves both — your frontend hits the GraphQL endpoint while design, PMs, and stakeholders see the data rendered as cards, tables, and forms. One source of truth, two audiences.

GraphQL endpoint For your code — with a project API key, from your server, CI or dev proxy 
{
  "data": {
    "users": [
      {
        "displayName": "Lena Hartmann",
        "role": "Designer",
        "verified": true
      },
      {
        "displayName": "Tomás Iglesias",
        "role": "Engineer",
        "verified": true
      },
      {
        "displayName": "Yuki Tanabe",
        "role": "Product",
        "verified": false
      }
    ]
  }
}
LH
Lena Hartmann
@lena.h Designer
TI
Tomás Iglesias
@tomas Engineer
YT
Yuki Tanabe
@yuki Product
POST https://api.mockomat.com/mock/abc-123/graphql
Rendered preview For everyone else — see the data, no Postman needed
LH
Lena Hartmann
@lena.h Designer
TI
Tomás Iglesias
@tomas Engineer
YT
Yuki Tanabe
@yuki Product
GET app.mockomat.com/s/abc-123/preview
05 Multi-tenant by default

Workspaces for the team.
Sessions for the moment.

Persistent workspaces hold your real schemas and roles. Disposable 24-hour sessions let you spin up a sandbox to share a link with a stakeholder — no signup, no cleanup.

Per-workspace roles & permissions Maintainer, editor, reporter, readonly · invite by email from Professional
Disposable sandbox sessions 24h from session start · no signup · auto-expires · sign up in the same browser to keep it
Isolated per tenant Models, members and API keys never cross accounts
Workspaces 3 active
N
Northvale Studio 12 members
ACTIVE
A
Acme Inc. 5 members
L
Lena · Personal 1 member
Sandbox session expires in 23h 12m · share via link
06 AI integration

Mockomat is also a prompt.

Tell Claude, Cursor, or any LLM to spin up a mock GraphQL API. They build it, you watch — like a movie. The whole team sees the schema take shape on the same board, in real time. By the time the model finishes its sentence, the link is in your clipboard.

Prompt Tell any LLM to use Mockomat.
MCP Native tool-use in Claude, Cursor, Copilot — coming soon.
OpenAPI Drop a spec, get a working mock back.
See the full AI integration story
claude · mockomat-mcp · session
LIVE
Reply to Claude…
Watching →
POARDVC
PM, architect, dev — and Claude — on the same board.
Product owner sees the data shape stakeholders need
Architect reviews relations + cardinalities live
Developer queries the mock the moment it lands
07 Under the hood

Schema-first. Stateless replay.
Yours to inspect.

Your model compiles to a deterministic GraphQL schema. Each request replays from the seed — same input, same output — so your tests stay reliable.

Read the architecture
schema.graphql deterministic runtime stateless · seedable /graphql JSON /preview Rendered UI
08 Start somewhere

Blueprints to start from, not stare at.

6 curated schemas modeled the way real products are. Fork one, rename, and you're ahead of every empty-canvas tutorial.

Users & Profiles Organizations, teams, user accounts, profiles, sessions, and audit trail
E-Commerce Products, customers, orders, and reviews for online shops
Cars & Vehicles Vehicles, manufacturers, dealers, listings, and test drives
Real Estate Property listings, agencies, agents, neighborhoods, and viewings
Furniture & Home Furniture catalog, designers, room designs, and customer reviews
Food & Restaurants Food products, stock management, nutritional data, and suppliers
Browse blueprints
09 Pricing

Start free. Pay when your team starts shipping.

Paid plans are for business customers only. Prices in EUR, net (excl. VAT), billed monthly.

Starter Free For individual developers
  • 1 project
  • 1,000 API requests / day
  • 2 guest collaborators per project
  • AI integration
RECOMMENDED Professional €19 / month, excl. VAT
  • 10 projects
  • Private endpoints & API keys
  • 11 team members
  • 50,000 API requests / month
Business €49 / month, excl. VAT
  • 20 projects
  • 250,000 API requests / month
  • 19 team members · 3 maintainers
  • Endpoint analytics
Compare every feature →
10 Questions

Things people ask us.

Can't find yours?Get in touch

Is this just another mock-API tool?
No. Mockomat starts from a collaborative ER model — schemas you can co-edit, comment on, and version. The mock API is the byproduct, not the interface.
GraphQL only, or REST too?
GraphQL today. The model compiles to a strongly-typed schema with pagination, filters, and relationships. REST adapters are on the roadmap.
How realistic is the data?
It comes from imported datasets: open data where a vetted source exists, generated sample data for the blueprints. The mock API serves those rows, or coherent combinations of their values — it invents nothing at request time.
Can I self-host?
No. Mockomat runs as a hosted service — there is no self-hosted version today.
What happens to a sandbox session?
It runs for 24 hours from its start — the app starts it when you first open it, so a project you create later has less time left. The hourly cleanup then deletes the project, which can take up to an hour. Share its collaboration link to edit it together. Sign up in the same browser to keep it as a project in your account; a project an AI built in its own guest session cannot be moved into an account.
Does it support custom auth & roles in the API?
Not per query. Your project's mock API checks an API key of the project (Professional and up); an optional actor token tells your consumers apart. Workspace roles control who edits the model, not what the mock API returns.