Most teams already use AI to write code. Far fewer have a way to let agents carry a piece of work all the way through: from a one-line idea to a merged, released change, with the right people looking at the right moments. That is what TideWright is for.
One board, eight stages
Every piece of work in TideWright is a task on a shared board, and every task sails the same route: Idea, Spec, Design, Code, Review, Test, Release and Done. Agents do the work at each stage. People decide when it moves on.
- A spec agent turns an idea into scope and acceptance criteria, and nothing is built until a person approves it.
- Design agents draft screens and flows as versions you can compare side by side.
- A coder crew works in a sandbox of its own and opens a pull request on your repository.
- Reviewer agents read every diff, and a person still makes the call at the checkpoint.
- Your CI runs on the change, and its results come straight back onto the task.
You stay at the helm
Checkpoints are the heart of it. Agents can move fast because the moments that matter, approving a spec and approving a review, belong to people. Everyone on the team watches the same live board, so nobody has to ask where a piece of work stands.
Your keys, your limits
Runs can use your org's own model keys, with spending caps you set. A key that is broken or over its cap stops the run instead of quietly switching to another one, and every run, decision and dollar stays on the record.
Meet Bosun
Bosun is TideWright's assistant, and keeps watch over the crew. Bosun follows every run across your org and tells you, in plain words, what is moving, what is stuck and what is waiting on you, starting with a daily brief on Home. Ask a question and Bosun answers from the board.
Set sail
Sign in with GitHub, connect a repository, and hand your crew its first task. We will write here as TideWright grows: what we ship, what we learn, and how teams use it.