Mano is a planning framework for AI-assisted development. Instead of asking an agent to generate a full project plan up front, it keeps work in small phases: scope the next phase, tighten only the artifacts that matter, break it into stories, implement one story at a time, and review before moving on.

I built Mano around the idea that human judgment should stay in control. It installs into a project as skills, templates, hooks, and an AGENTS.md contract, giving AI IDEs a shared workflow to follow without pretending to be autonomous planning systems.

Mano encourages short feedback loops, explicit approval of phase scope, and living artifacts that evolve as real implementation feedback invalidates earlier assumptions. Install it with npx mano-plan install, then use commands like mano start, mano stories, mano dev, and mano review from your AI IDE’s chat to guide a project from rough idea to supervised implementation.