EvoMaster - AI Developer Tools Tool

Overview

EvoMaster is an open-source, research-driven fuzzer that automatically generates system-level test cases for web and enterprise APIs (REST, GraphQL and RPC). It combines evolutionary algorithms with dynamic program analysis and, when possible, white-box bytecode inspection to evolve test suites that maximize coverage and reveal faults; outputs are executable tests (JUnit for Java/Kotlin, plus Python and JavaScript formats) that can be used for regression testing. According to the project repository, EvoMaster produces self-contained tests (white-box mode can start/stop the SUT), creates an interactive HTML web report by default, and can be run fully in-house without external telemetry or paid LLM services. ([github.com](https://github.com/WebFuzzing/EvoMaster)) EvoMaster is actively used in both academia and industry: peer-reviewed evaluations and user studies report it among the top-performing REST/GraphQL fuzzers (comparative studies and an industry case study at Meituan), and the project provides Docker images, platform installers and an “uber” executable JAR for offline use. The tool is designed for longer runs (recommended 1–24 hours for best results) but also provides quick-start Docker commands for short experiments. EvoMaster is distributed under LGPL-3.0 and funded by research grants. ([link.springer.com](https://link.springer.com/article/10.1007/s10515-024-00478-1?utm_source=openai))

GitHub Statistics

• Stars: 674
• Forks: 103
• Contributors: 23
• License: LGPL-3.0
• Primary Language: Kotlin
• Last Updated: 2026-02-25T08:39:35Z
• Latest Release: v5.0.2

The GitHub repository shows sustained maintenance and active development: the project has several hundred stars and forks (674 stars, 103 forks) and a dozen-plus active contributors; commits are recent (multiple commits in February 2026), indicating ongoing fixes and feature work. The repo is licensed LGPL-3.0 and provides release assets (including an uber JAR and installers) and Docker images. Open issues and active pull requests demonstrate an engaged user/developer base and ongoing feature requests (GraphQL improvements, SQL handling, security oracles). Overall community health is strong for an academic-driven OSS project with industrial adoption. ([github.com](https://github.com/WebFuzzing/EvoMaster))

Installation

Install via docker:

docker pull webfuzzing/evomaster

docker run -v "$(pwd)/generated_tests":/generated_tests webfuzzing/evomaster --blackBox true --maxTime 30s --ratePerMinute 60 --bbSwaggerUrl https://petstore.swagger.io/v2/swagger.json

java -jar evomaster-<version>-uber.jar --help  # (project publishes an 'uber' executable JAR; see Releases for exact filename)

Key Features

• White-box JVM analysis: bytecode instrumentation, taint analysis and testability transformations to improve test effectiveness.
• Black-box mode: fuzz any API reachable over HTTP using OpenAPI/Swagger or GraphQL introspection.
• Executable test output: generates JUnit (Java/Kotlin), Python and JavaScript test suites suitable for CI/regression.
• Database-aware fuzzing: inspects SQL/MongoDB interactions and can seed Postgres, MySQL, H2 or MongoDB for tests.
• Interactive report + CI: produces an index.html web report and offers a GitHub Action for CI integration.

Community

EvoMaster is research-backed and used in industry (academic papers and multi-year Meituan studies report clear advantages in coverage/fault detection). The GitHub repository shows active commit cadence and open collaboration (issues, pull requests, Discussions). Funding is provided by ERC and other research grants, and release artefacts (Docker images, installers, Zenodo release snapshots) support reproducible use. Users raise practical requests in issues (GraphQL edge cases, SSL/test flakiness, SQL handling), indicating both effective adoption and an active roadmap. ([sciencedirect.com](https://www.sciencedirect.com/science/article/pii/S0167642325000619?utm_source=openai))

Key Information

• Category: Developer Tools
• Type: AI Developer Tools Tool