APIEval-20 is a black-box benchmark for API testing agents. Each agent gets only a JSON schema and one sample payload, then generates a test suite. We run those tests against live reference APIs with planted bugs and score bug detection, API coverage, and efficiency. Unlike LLM-as-judge evals, scoring is fully objective: a bug is either caught or it isn’t. Tasks span auth, errors, pagination, schemas, and multi-step flows. Open on Hugging Face.

Open-source TUI for Playwright testing. Record your flow in the browser, then everything happens in your terminal. No tab-switching to ChatGPT/Claude, no copy-pasting, no manual context juggling. Bring your own API key (OpenAI, Claude, Gemini). Runs entirely local. Our LLM orchestration expands one recording into comprehensive test coverage - edge cases, error handling, boundary conditions - more efficiently than calling LLMs directly. Record, generate, run: all terminal-native. MIT licensed.

Kusho is an AI Agent for API testing. It generates an exhaustive test suite for your API in 2 minutes– just put in an API spec and get a host of functional tests covering real-world scenarios. Run these tests with AI-generated assertions in a single-click.

An open-source AI Agent that lives in your terminal. Generate UI tests with one command, run them instantly or plug into your CI pipeline. No boilerplate, no brittle scripts — just reliable testing built by developers, for developers.

KushoAI finds bugs in backend user journeys. Allow KushoAI to run contract testing for your APIs- just add API specifications and watch as an exhaustive test suite gets generated from scratch. Execute these tests with auto-generated response validations.

KushoAI helps you test web user journeys in minutes. Record user journeys using the KushoAI extension and watch as exhaustive test code gets generated. Find bugs without having to think of scenarios and write endless automation scripts to run them.