Caffeine Projects

Mock HTTP and gRPC servers for Go tests, and a CLI that serves them from files.

VLatte starts real servers that answer the way you tell them to. Start one in a Go test, or run the vlatte CLI to serve mocks from YAML or JSON files.

• HTTP and gRPC, with no gRPC service implementation to write: unary and every kind of streaming call.
• Matchers on any part of a request: JSON path, JSON Schema, XPath, regex, globs and more.
• Templated replies, sequences, scenarios for stateful flows, and webhooks.
• Proxying and recording of a real service to play it back later.
• Fault injection: latency, slow bandwidth, connection resets and malformed answers.
• An admin API to manage mocks while a server runs, also served over MCP.
• A CLI and a Docker image to run mock servers outside Go.

go get github.com/caffeine-projects/vlatte

brew install caffeine-projects/tap/vlatte

go install github.com/caffeine-projects/vlatte/cmd/vlatte@latest

docker pull ghcr.io/caffeine-projects/vlatte

Documentation · Repository · Go reference