Everything you need to get set up - and answers to the questions we hear most.
The one thing to understand: Yeeted is driven by your coding agent. You connect the agent once (it takes a minute), and from then on you deploy by asking - "Deploy this project to Yeeted" - not by filling in forms. The getting-started walkthrough covers this step by step; the short version:
1. Copy your MCP URL: loading…
2. Add it to your agent as an MCP server (per-agent instructions are in the walkthrough and on the Credentials page).
3. First use opens a browser window - log in and approve. Done.
Using Claude? The easiest path is Claude's connector settings - visit claude.ai/new#settings/customize-connectors, add a custom connector, and paste your MCP URL. Claude Code users can instead run:
loading…
Open a project in your agent and ask it to deploy:
"Deploy this project to Yeeted."The agent packages your code, builds it, and replies with a live URL. Your app then appears on the dashboard with its status, logs, and metrics.
Three usual causes: the MCP server wasn't added (re-check the config against the walkthrough), the agent wasn't restarted after adding it, or the OAuth authorization was dismissed - remove and re-add the server to trigger the browser approval again.
Every new account starts on the free plan automatically - there's nothing to choose at signup. It includes 5 projects and a budget of 5 deployments and builds, which does not reset. When you run out - or want a managed database, always-on production, more memory or multi-instance - upgrade from Billing (Developer is $10/mo). Cancelling any paid plan puts you back on the free plan.
No - API keys don't exist on Yeeted. Interactive agents authorize through OAuth in the browser (no credentials to copy). For CI jobs and scripts, create an OAuth credential (client id + secret) on the Credentials page; secrets are shown once, at creation.
Yes - just ask: "Deploy this app with a Postgres database attached." Postgres, MySQL, MariaDB, MongoDB, and Redis are supported. The connection URL is injected into your app's environment automatically; credentials live in the secrets vault, and you can export a dump of your data any time.
Development apps and their databases idle-stop to save resources - they wake on the next deploy or request. Production apps (Business and up) are kept always-on, restarted on crash, and load-balanced across servers.
Every deploy is kept with its build. Ask your agent to roll back — production releases switch blue/green, so the previous version comes back without a rebuild. Production databases are snapshotted before risky operations.
Yes. Add domain: yourdomain.com to your project's yeeted.yaml
and deploy, then finish setup on the Domains page — it shows you exactly what
to point where, and Yeeted obtains and renews a free Let's Encrypt TLS certificate automatically once
your DNS points at us.
Your account is homed to a geographic area at signup (e.g. US or EU), and everything you deploy — apps, databases, logs — stays inside it. Data is never moved or replicated across areas. You can download everything we hold about you from your account settings, and closing your account permanently erases your personal data after a 30-day grace window.
Credentials page → find the client → revoke. The agent loses access immediately for new tokens; in-flight sessions end within a minute. Re-connecting later just repeats the one-time browser approval.
Start with your app's logs and build output on the dashboard (your agent can also fetch them — ask "show me the logs for my app"). Build failures include the compiler or tooling error verbatim. If a deploy sits in "deploying", the build log usually says why.
Still stuck? Ask your agent — it can read your deploy status, logs, and build errors directly. This is a research preview: things move fast, and your data may be reset between releases.