Run Your App
Run configurations are named groups of commands that launch together in parallel sub-tabs. Instead of opening separate terminals to start your API server, frontend dev server, and database, you define a run config once and launch everything with one click from the dashboard.
What you’ll need
Section titled “What you’ll need”- A repository with Cate set up (quickstart)
- The shell commands that start your project (for example,
npm run dev,docker compose up)
Start the interview
Section titled “Start the interview”Open the Run Config command from the dashboard.
If you already have run configurations in .cate/run-configs.json, the agent shows your current configurations and asks whether you want to modify them, add a new one, or start fresh.
If no configurations exist, the agent begins the interview.
The interview
Section titled “The interview”The agent walks through a short set of questions. It auto-detects common patterns from your package.json, docker-compose.yml, or build files and skips questions it can answer itself.
How do you run your project? Is it a single command like npm start, or do you need multiple things running at once — an API server, a frontend dev server, a database?
What are the commands? For each thing that needs to run, the agent asks for the shell command and a short name for the sub-tab label. For example: “Database” runs docker compose up, “API” runs ./gradlew bootRun, “Frontend” runs npm run dev.
What’s the configuration called? This appears in the launcher dropdown. The agent suggests a name: “Dev Server” for a single command, “Full Stack Dev” for a multi-service setup.
Do you need more than one configuration? If you have different setups for different situations — development versus staging, lightweight versus full — the agent repeats the questions for each additional configuration. Each becomes a separate entry in the launcher dropdown.
Before generating, the agent presents a summary of what will be created and confirms it’s right.
Examples
Section titled “Examples”Single command
Section titled “Single command”A project that runs with one command:
When you launch “Dev Server,” one sub-tab opens running npm run dev. Even for a single command, a run config is useful — it gives you a named, one-click launcher without opening a terminal.
Multi-command parallel
Section titled “Multi-command parallel”A full-stack Node.js and React project that needs three things running simultaneously. When you launch “Full Stack Dev,” three sub-tabs open at once — “Database” running docker compose up, “API” running npm run dev:api, “Frontend” running npm run dev:web. Each sub-tab name appears in the tab bar so you can monitor each service independently.
Multiple configurations
Section titled “Multiple configurations”A project with development and staging environments. “Development” opens two sub-tabs with dev API and web commands. “Staging” opens two sub-tabs with staging commands. You pick which one to launch from the dropdown.
Save and share
Section titled “Save and share”After generating, the agent writes the configuration to .cate/run-configs.json and validates the structure. You then choose how to save it:
- Branch and PR — the agent creates a branch, commits, pushes, and opens a PR. Useful for sharing the configuration with your team.
- Commit directly — the agent commits to the current branch without a PR.
- Local only — the file is written but not committed. You handle it yourself.
New run configurations appear in the launcher after the changes are merged and you re-open the repository in Cate.
See Run configs reference for the full JSON schema and field descriptions.