1. Create your workspace
Sign up at app.onepatch.dev. You authenticate through WorkOS, then create an organization — give it a name and confirm your domain. Creating the organization starts provisioning your workspace: a private OnePatch instance with its own server and AI agent. Provisioning runs on its own and usually takes a few minutes. The setup screen shows live progress, and the onboarding chat tells you when your workspace is online. You can read the rest of this guide while you wait.A workspace is yours alone. Your telemetry, your repositories, and your agent’s work all live inside it and are not shared across organizations.
2. Connect GitHub
The agent reads and writes your repositories — it explores your code to understand your services, and it opens pull requests when it has a fix. Connecting GitHub has three quick steps, shown in order by the onboarding chat:Authorize OnePatch
Click Connect GitHub. This opens an OAuth popup and grants OnePatch access to your account.
Install the GitHub App
Install the OnePatch GitHub App on the account that owns the repositories you want to use.
3. Connect Slack
Click Connect Slack and authorize the workspace. OnePatch creates an#onepatch-alerts channel and posts there when a monitor fires. This is where incident notifications land, so connect the Slack workspace your team actually watches.
4. Send telemetry from your app
OnePatch reads OpenTelemetry. To send it, you instrument your application once. The onboarding chat hands you a single block to paste into your coding agent — Claude Code, Cursor, or similar — which installs a setup skill and wires up the OpenTelemetry SDK for you. The block looks like this, with your own endpoint and token filled in:http/protobuf (the OpenTelemetry SDK default) and http/json work, as does OTLP over gRPC. You do not need to change your exporter’s protocol; the setup skill configures it for you.
The setup skill is open source. You can read exactly what it does at github.com/1patch/skills before you run it.
After the skill finishes, commit and push the changes. Your app starts sending telemetry the next time it runs.
5. Get your first dashboard
When your first telemetry arrives, the agent reads theTELEMETRY.md file the setup skill wrote at your repository root — a short description of what your service emits — and drafts a first dashboard from it. It appears in the side pane next to the chat and refreshes as more data lands.
From here, you change it by asking. For example:
Where to go next
Your workspace is now set up. The rest of the app is available from the left navigation:Chats
Talk to the agent. Ask it to build dashboards and monitors, investigate your data, explore your code, or open a pull request.
Dashboards
Browse the dashboards the agent has authored. They update live as telemetry arrives.
Monitors
See every monitor, its schedule, and its current state.
Integrations
Manage connected repositories and Slack, and copy your telemetry endpoint and token.