Fault injection

Introduce controlled failures into integration tests.

Expires after 1 hour

Start with routes or import an OpenAPI document to build your mock.

Checking this session…

The example below already works.Create it now and edit it later, or change it first.

Proxy & record a real API Pro

Point the mock at your real API: unmatched calls pass through and real answers are saved as routes. Beeceptor charges $10 a month for its paid plans; this is part of Kivel Pro.

Get Pro for $2 a month

Import / export configuration

Mock configuration (JSON)

API name

MethodPathStatusDelay (ms)Requests per minuteRandom failures (%)DeliveryResponse type

Request conditions & response templates

Match request fieldResponse mode

Templates accept {{path.id}}, {{query.name}}, {{body.name}} and sample data such as {{fake.name}}, {{fake.email}}, {{fake.uuid}}, {{fake.price}} or {{fake.datetime}} inside JSON strings or plain text. Missing values become empty strings. No code runs. Test credentials are stored in the configuration; use fake tokens only.

Response body

Up to 20 routes, 80 KiB total configuration, and 120 requests per minute. Paths support parameters such as /users/:id. Matching routes are checked in order.

Open an existing mock API

Resource IDManagement token (optional for your workspace)

About Fault injection

Create a test endpoint with a controlled failure status and optional response delay.

When to use it

Create a test endpoint with a controlled failure status and optional response delay.

Fault injection example

GET /hello → HTTP 503

How to use it

Configure the endpoint and choose private or public access.
Create it and send requests from your client.
Inspect test output, update or delete the resource when finished.

Limits & privacy

Uses an expiring mock resource, private by default, with separate management and access credentials. Public access is optional. Standard resource quotas, ownership, renewal and deletion apply. Private access tokens are sent in headers, never in share URLs. Static failure responses with statuses 200–599 and delays up to two seconds. No packet loss or low-level transport fault injection.

Use this tool from the API reference or connect through the SDK, CLI, SSH, or MCP. Hosted calls share your plan allowances.