Keploy

Keploy converts live application traffic into reproducible tests and dependency sandboxes, without touching your application code. It uses eBPF and network-level interception to capture incoming requests alongside every outgoing call to databases, queues, caches, and external APIs, then packages all of that into editable tests and mocks you can replay against any new build.

The core use case is regression detection. When a developer (or an AI coding agent) changes something, Keploy replays the recorded behavior and flags differences in responses, schemas, contracts, or performance. That loop runs locally or inside a CI/CD pipeline, so problems surface before code is merged.

Key capabilities:

• Record and replay real traffic, including dependency state, without writing a single mock by hand
• AI-assisted test generation that extends recorded coverage with edge cases, connected flows, and schema-driven scenarios
• Deduplication to keep test suites lean rather than accumulating redundant cases from high-volume traffic
• API and schema coverage reporting so you can see what's actually exercised
• Kubernetes support in the enterprise tier for recording from staging or production clusters, with governance dashboards and security controls

Compared to tools like Mockoon for manual mock creation or Testfully for request-based API testing, Keploy's distinguishing move is deriving both tests and mocks from real behavior rather than requiring developers to define and maintain them. That matters most for teams where integration test coverage is thin because writing mocks is slow, or where AI-generated code is being merged faster than manual tests can keep up.

The core record-and-replay platform is open source under Apache 2.0. Enterprise features add production traffic recording, multi-environment support, and team governance on top.