Maggy is an MIT-licensed local AI engineering harness that routes coding tasks across Claude Code, Codex, Kimi, and Gemini CLI. It provides test and quality gates, cross-session memory, isolated parallel agents, MCP-based code graphs, plugins, and an optional local dashboard.
In the Frameworks & runtimes space, Maggy takes a focused approach. It focuses on running governed, test-enforced multi-model coding workflows locally without relying on a hosted gateway. It is built as an open-source project for software developers and AI engineering teams. Maggy is open source under the MIT license. Maggy is available on the web, the command line, and API, and it can be self-hosted.
Alinaqi builds and maintains Maggy, and it first shipped in 2025. The project is developed in the open on GitHub with 704 stars and 90 commits in the last 90 days. Key capabilities include multi-model routing, quality gates, and test enforcement. The interface is available in Arabic, German, and Japanese. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do