YYLO
YYLO (pronounced why-lo) is an open-source command-line orchestrator designed to manage AI coding agents in software development workflows. It provides a structured pipeline where each task is executed in a dedicated Git branch and worktree, ensuring isolation between concurrent agent activities. The tool introduces typed task boundaries, validation steps, merge queues, and release-readiness checks to maintain code quality and prevent conflicts. The orchestrator operates through equivalent `yylo` and `yy` launchers, offering flexibility in how developers invoke it. By enforcing a risk-based merge queue review, YYLO helps teams evaluate changes based on their potential impact before integration. This approach is particularly useful for teams that rely on multiple AI agents to generate or modify code, as it brings governance and control to the automated development process. As an open-source project, YYLO can be self-hosted and customized to fit existing Git-based workflows. It targets developers and teams looking to scale AI-assisted coding without sacrificing codebase stability. The tool's emphasis on task boundaries and release readiness makes it suitable for both experimental projects and production environments.
- Category
- AI Agents
- Pricing
- Open Source
Problem solved
YYLO solves the coordination problem of running multiple AI coding agents on the same repository. Without orchestration, agents can interfere with each other, produce conflicting changes, and introduce unreviewed code. YYLO isolates each task in its own branch and worktree, applies validation and merge policies, and uses a risk-based review queue to ensure only safe, release-ready changes are integrated.
Use cases
- Running multiple AI coding agents in parallel on separate features
- Enforcing code review and validation on AI-generated changes
- Managing a queue of agent tasks with risk-based merge prioritization
- Preparing codebases for release with readiness checks
- Integrating AI agents into existing Git branch workflows
- Automating repetitive coding tasks with safe rollback
Audience
- Developers
- Startups
- Enterprise
Pros
- Isolates agent tasks in dedicated branches and worktrees
- Risk-based merge queue improves code review efficiency
- Typed task boundaries enforce structured workflows
- Open source and customizable
- Supports release-readiness validation
Cons
- Requires Git and command-line familiarity
- May add overhead for small or single-agent projects
- Opinionated workflow may not fit all teams
- Documentation and community support may be limited
Official website: https://github.com/yylo-dev/yylo
Canonical LambdaBase page: https://www.lambdabase.com/tools/yylo