Questions, answered directly
Including the ones where the honest answer is “not yet.” The guides go deeper.
What it is
- What is Arcadia Mission Control?
- A governance and prioritization layer for AI-assisted software production. It holds what you are building and why, decides which work comes next, stops at the decisions that should stay human, and verifies that approved work actually completed. It sits above your coding agents rather than replacing them.
- How is this different from a coding agent?
- Coding agents write code. Mission Control does not write any. It decides what the agents should work on, keeps their sessions and branches from colliding, and checks the result against criteria written before the work started. Different job, different layer.
- How is this different from an orchestration framework?
- Orchestration frameworks route and parallelize execution — they make more agents run at once. Mission Control governs intent, approval, priority, and evidence. It answers what should run and whether it worked, not how to run more of it.
- How is this different from an issue tracker?
- A tracker stores intent but never decides. It will not tell you what to do next, will not stop an agent at an approval boundary, and will not verify that a closed ticket reflects work that happened. Mission Control does those three things.
- Does “verified” mean it looks and feels right, or just that it runs?
- It means the work meets criteria written down before it started — a page that renders, a flow that completes, a test that passes. Written that way, criteria can capture a lot of what good UX cares about, but they are not a judgment of taste. Whether it feels right is still your call.
Running agents
- Which coding agents does it work with?
- Arcadia currently drives Claude Code, Codex, and OpenCode. Coding agents remain interchangeable execution tools under your direction.
- Will Arcadia make product or design calls without me?
- No — that is the one thing it is built not to do. Anything consequential is a decision that stops and waits for a person: what ships, what does not, anything that cannot easily be undone. Arcadia decides what runs next inside the boundaries you set; it does not set the boundaries.
- How much autonomy does it actually take?
- As much or as little as you set. The design intent is that consequential actions — spending money, deploying, anything irreversible — stop and wait for a person, while ordinary work continues. How much autonomy people actually want is an open question we are asking directly.
Fit and availability
- Is it available yet?
- Arcadia runs as the system used to build itself. Early access opens when someone who is not the author can set it up from cold and have it work.
- What does it cost?
- Nothing is for sale, and there is no price yet. Nobody has paid for Arcadia; early-user feedback helps shape that decision.
- Does it need access to my code or my repositories?
- The website collects no code, repository access, or credentials — only what you type into the form. Arcadia runs locally, on your machine and your repositories. A hosted option is being explored.
- Who is building this?
- MartianRover — a very small operation, currently one person. That is a reason to be skeptical about longevity and a reason it can change direction fast based on what early users say. You can reach a human at [email protected].