Snow-White answers the question every engineering team asks: “Which API endpoints are our tests actually exercising?” It correlates the OpenAPI specs you already have with the OpenTelemetry traces your applications already emit — no proprietary agents, no code changes.

That question matters more than ever in the agentic era. When an AI coding agent writes both the implementation and the tests, a green test suite no longer proves the two agree with your API contract — only with each other. Snow-White checks real traffic against the real spec, an independent, ground-truth guardrail no agent can talk its way around.

API Coverage Dashboard

Coverage at a Glance

See exactly which endpoints, parameters, and responses are exercised by your test suite or production traffic — endpoint by endpoint.

Start Measuring

Quality Gate Results

Automated Quality Gates

Define coverage thresholds and validate them in CI/CD. The CLI exits non-zero when coverage drops — catch regressions before they ship.

See Pipeline Workflows

Architecture Overview

Cloud-Native by Design

Deploy with a single Helm command. Event-driven microservices on Kafka, built for Kubernetes with configurable scaling.

Explore Architecture


How It Works

📥
1. Sync Specs

Import OpenAPI specs from Artifactory. Snow-White keeps them in sync automatically.

📡
2. Ingest Traces

Applications emit OTEL traces during tests or in production. No code changes needed.

🔗
3. Correlate

Traces match specs via service name, API name, and version. Coverage computed automatically.

🚦
4. Gate

Evaluate results against quality thresholds. Fail fast when coverage regresses.

📊
5. Visualize

Review reports in the UI or export results to CI/CD via the CLI.

OpenTelemetry Integration

OpenTelemetry Native

No proprietary SDKs. Snow-White piggybacks on the OTEL traces your services already emit. The Spring Boot autoconfiguration and OpenAPI Generator plugin wire the three linking attributes automatically.

openapi: 3.1.2 info:
  title: My Service API
  version: 1.0.0
  x-api-name: my-api
  x-service-name: my-service

Instrument Your Service

Agentic mode JSON summary

Built for Agentic Development

snow-white calculate --agentic emits a single JSON summary naming exactly which paths, parameters, and response codes are undercovered — a concrete backlog a coding agent can act on directly. Pair it with the Claude Code skill for fixes proposed automatically.

See the Claude Code Skill


Quick Start

helm repo add snow-white https://bbortt.github.io/snow-white
helm repo update

helm install my-snow-white snow-white/snow-white \
  --set snowWhite.host=snow-white.example.com

For full configuration options see the Deployment Guide.