Maige uses AI to label issues in repositories, simplifying task management.
Maige is an open-source infrastructure for running natural language workflows on your codebase. It connects to your repository, builds embeddings of the entire codebase, provides a sandbox environment, and lets you describe automated workflows that operate on issues, PRs, and code. It integrates with the GitHub API to perform tasks such as labeling, assigning, commenting, reviewing code, and even running simple code snippets. The system is designed to be configurable via textual rules and supports an AI-assisted approach to code review and governance. Pricing is usage-based, with a free tier and enterprise options, and the project emphasizes an open-source, experiment-first approach.
Maige flexibly interacts with the GitHub API to find a way to automate common maintenance and review tasks. It can auto-label, auto-assign, auto-comment, and perform code reviews, with a code sandbox that it can spin up. The system is highly customizable via textual instructions and can also generate or execute code snippets as part of its automation.