Project description
TranslatedWe're building a new web product and the frontend team is idle waiting for a backend that won't be ready for several more weeks. We need a dedicated mock server that reads our OpenAPI specification and takes the place of the real service, serving predictable pre-made responses for each described endpoint. The goal is to get a stable contract that client applications can be developed and demonstrated against right now, so when the real backend arrives, switching is just a matter of changing the base URL. This is specifically a mocking and stub tool, not an API gateway or test framework: routing, authorization proxying and assertion libraries aren't needed, just authentic fake responses.
The mock should parse the specification and automatically generate payload examples from declared schemas, but also allow manually overriding any route with custom response body bound to path, method and status code. I want to simulate real conditions on demand: configurable delay for each endpoint, forced error codes like 400, 401, 404, 429 and 500, plus the ability to return different responses based on a request header or flag so the team can test both happy paths and error handling. Response fixtures should live in plain files that we store in git, and reloading them shouldn't require a restart. A small config can set global default values for delay and error rates.
On output I expect a single service that starts locally with one command and fits the same way in a container for a shared environment. Include clear instructions and a couple of ready scenarios.
— Parse OpenAPI specification and generate response examples for each endpoint
— Override responses by path, method and status through editable fixture files
— Simulate delays, timeouts and forced error codes on demand
— Hot reload fixtures without restarting the server
— Run locally with one command plus container image for staging
The mock should parse the specification and automatically generate payload examples from declared schemas, but also allow manually overriding any route with custom response body bound to path, method and status code. I want to simulate real conditions on demand: configurable delay for each endpoint, forced error codes like 400, 401, 404, 429 and 500, plus the ability to return different responses based on a request header or flag so the team can test both happy paths and error handling. Response fixtures should live in plain files that we store in git, and reloading them shouldn't require a restart. A small config can set global default values for delay and error rates.
On output I expect a single service that starts locally with one command and fits the same way in a container for a shared environment. Include clear instructions and a couple of ready scenarios.
— Parse OpenAPI specification and generate response examples for each endpoint
— Override responses by path, method and status through editable fixture files
— Simulate delays, timeouts and forced error codes on demand
— Hot reload fixtures without restarting the server
— Run locally with one command plus container image for staging
Texts and translation Cash, Bank transfer
