Admins can set workspace-wide instructions under Settings → Agent — your team’s conventions, production context, and rules for taking action.
The Services page brings together the agent’s map of your system: each service’s health, monitors, dashboards, and dependencies. See what calls what, including the external services your system relies on.
Collect product analytics, crash logs, and other telemetry from native apps built with React Native and Expo. The agent investigates mobile issues alongside your backend telemetry.
Open the charts and tables behind the agent’s conclusions directly from an incident report to inspect the data yourself.
Claude Code, Codex, and Cursor can query your telemetry, read incident investigations, and hand work to the OnePatch agent from inside the editor.
npm install -g onepatch && onepatch install sets up every agent on the machine. See Coding agents.Connect Linear from the Integrations page and the agent reads your issues for
context while it investigates, and can create and update them in your workspace.
It works under its own Linear identity, so everything it writes is attributed
to OnePatch.
⌘K finds anything in the workspace — chats, incidents, monitors, and dashboards. The agent searches the same index, so a new incident starts with a look through past ones.
The agent can use the internet while it investigates, scoped to a domain allowlist you control. See Network allowlist.
After a pull request merges, the agent watches the change in production and reports on the PR whether it held up. A review comment before merge says what it will watch, and a regression opens an incident. See Verify PRs in prod.
Point the agent at a Slack channel and it reads every new message there, opening a thread on the ones that need attention. Turn it on from the integrations screen or with
@onepatch watch in the channel. See Watched channels.When something breaks for a user, the agent can see what that user actually did — the pages, clicks, failed requests, and JS errors of their session, each joined to the backend trace it caused. Setup is adding
@onepatch/rum to the frontend. See Browser monitoring.The home page leads with the pull requests the agent has opened, each with the monitor history that prompted it — enough to judge the change without leaving the page. Underneath, a table lists every pull request across your connected repos.
The app lays out for small screens instead of assuming a desktop window.
When a new incident turns out to be part of one already open, the agent merges them into a single episode. You follow one thread instead of a row of separate chats.
Connect an IAM role from the Integrations page and the agent can describe resources, read logs, and take whatever actions the role’s policies allow while it investigates. Permissions stay on your side, and every action lands in your CloudTrail named for the chat that made it. See AWS.
One workspace-wide memory that carries across chats, so the agent doesn’t start cold every time.
Every workspace gets an ingest address. Forward alert email from any vendor and it becomes telemetry a monitor can act on.
Mint machine credentials for the MCP endpoint from the app, and rotate them without downtime.
A check going green re-opens the question instead of closing the incident on its own.
OTLP histograms are exposed as a typed view the agent can query directly.
Chat, dashboards, monitors, the command palette, and onboarding are navigable by keyboard and announced to screen readers.
Hand-off retries every tick until an agent acknowledges it, and resolution waits until the check is observed green. A dead-man alarm fires if monitor evaluation itself stops.
The Slack card rewrites itself as the incident moves through its phases, so the channel shows current state rather than a pile of updates.
Incident chats open with a status-page-style card — what fired, what phase it’s in, what changed — pinned beside the transcript.
Connect any OAuth-protected MCP server from the Integrations screen and the agent can use its tools. See How OnePatch works.
A full-page switcher lists every org you belong to.
Rows collapse to a status sparkline and open into a two-pane detail view.
DM the OnePatch bot, or @-mention it in a channel and it replies in a thread. Each conversation is a real chat that also shows up in the app, with the sender’s name. See Incidents and Slack.
The agent can define dropdowns and multi-selects that sit under the time tape and filter the whole board.
Every root
.tsx file in your workspace becomes a dashboard tab named after the file.Queries run over much longer windows and return faster. Your signals live in a ClickHouse database scoped to your workspace.
Every workspace runs in its own private network with signed images and no credentials shared between workspaces.
Public docs at docs.onepatch.dev, with a getting-started guide and reference for monitors, dashboards, and incidents. Written to be read by an agent as well as a person — the full text is at llms-full.txt.
A monitor that fires opens a chat named after it, and the agent works the incident from there through a real state machine. A check that can’t run at all pages the agent too.
A saved dashboard opens at full width on its own page. Axes and tooltips carry units, and dragging to zoom applies the new range to every chart on the board.
Fixes arrive as a PR against your connected repo rather than a suggestion in chat.
Org-wide chats where every message carries its author, so the agent knows who is speaking.
When the agent needs a decision mid-task, the choices render in the composer and your answer goes straight back into the run.
⌘K jumps to any chat, dashboard, or monitor.
A monitor is a TypeScript program on a cron that queries your telemetry and returns a result. The agent writes and edits them; they live in a folder tree in your workspace repo.
Send a message mid-response and the agent picks it up instead of making you wait for the turn to end.
Drag or paste a screenshot into the composer.
You get an OTLP endpoint and a token. Anything you send becomes queryable by the agent.
Sign in, connect a GitHub repo, get a provisioned workspace with an agent in it.