Skip to main content
Every project starts with a single development environment called default. A dev branch is an isolated environment that lives under development, with its own runs, schedules, and concurrency. Branches are useful when you run more than one local dev session at a time. Give each session its own branch so their runs don’t collide:
  • Run several git worktrees or copies of your project in parallel, one branch each.
  • Let multiple coding agents each work in their own branch without stepping on one another.
When you’re done with a branch, you can archive it to free up a slot or just re-use it.

​
Run a dev session on a branch

Log in with the CLI first:
Then start a dev session on a branch with the --branch flag. If the branch doesn’t exist yet, it’s created:
Without --branch, the session runs on the default branch.
You can also set the branch with the TRIGGER_DEV_BRANCH environment variable instead of the flag.

​
Archive a branch

Archive a branch from the CLI when you no longer need it. Pass the branch with --branch (or set TRIGGER_DEV_BRANCH); it’s required, and the default branch can’t be archived:
You can also create and archive branches from the Dev branches page in the dashboard.

​
Limits on active branches

Each branch has its own concurrency, so we limit how many dev branches each developer can have active per project. The default development environment counts toward this limit. If you are at the limit when you run dev --branch, the CLI automatically archives up to 3 of your branches that have no connected dev session and have been idle for over an hour, to make room. You can also archive a branch any time to free a slot. Need more? Get in touch and we’ll raise the limit.