# Quickstart: zero to dashboard > Sign in with GitHub, install the GitHub App on your organization, select repositories, and see your first CI analytics. Takes about five minutes. Source: https://latchkey.dev/documentation/quickstart ## Summary - Sign in with GitHub, install the app on your org, select repositories: about 5 minutes, and the 30-day trial starts on its own. - Backfill imports the last 30 days of completed runs, so dashboards fill within minutes; AI Insight findings follow nightly. - Optional first runner job: change one line to `runs-on: latchkey-small`. This is the fastest path from nothing to a working Latchkey dashboard. Nothing in it touches your repositories or workflows; the only code change on this page is the optional one-line `runs-on:` switch near the end. **Before you start** - [ ] I have a GitHub **organization** (personal accounts are not supported) - [ ] I have owner or admin permissions on that organization - [ ] I am on a desktop browser (the dashboard is desktop-only) 1. **Sign in with GitHub** Go to [latchkey.dev/sign-in](https://latchkey.dev/sign-in) and sign in with your GitHub account. Use the GitHub account that has access to the organization you want to monitor: the app install in the next step happens under this identity, and the person who installs the app becomes the workspace owner. You land on the dashboard's **Get started** card ("Welcome to Latchkey"), and your 30-day trial is already running, no credit card required. If you get interrupted at any point, just sign in again; setup resumes where you left off. 2. **Connect GitHub** Step 1 on the Get started card is **Connect GitHub**. Click it and Latchkey sends you to GitHub to install the app (a short interstitial explains what happens first). When GitHub asks where to install, choose your **organization** (not your personal account). You need owner or admin permissions on the organization; if you have neither, ask an organization owner to install the app or grant you admin rights. What the app can and cannot access is covered in [Install the GitHub App](/documentation/install-the-github-app). 3. **Select repositories to monitor** Step 2 on the Get started card, **Select repositories**, opens the repository picker, which lists your organization repositories with search and a visibility filter. Select the ones you want Latchkey to monitor; the counter shows how many your plan allows (10 on Launch, where every new workspace starts; 1 on Developer and 40 on Scale if you change plans later). Two things worth knowing while you choose: managed runners only serve monitored repositories, so include any repo you plan to point at `latchkey-*` labels; and the "Skip for now, explore the dashboard first" link lets you defer this and add repositories later from the dashboard. Guidance on choosing under a plan limit is in [Managing repositories](/documentation/managing-repositories). 4. **Watch the dashboard fill in** After setup, a short guided tour of the dashboard is offered; you can skip it at any point. For each repository you enabled, Latchkey backfills recent completed workflow runs from the last 30 days, so most teams see data within a few minutes; a progress banner shows while backfill runs. If a page looks empty, check the filter bar at the top first: analytics pages show data for the repositories you select and prompt you to pick one if nothing is selected. 5. **Optional: run a job on a Latchkey runner** Change one line in any workflow to try a managed self-healing runner. After you push, the job shows as queued on GitHub until a runner picks it up: about two seconds when a warm runner is already online, longer when a fresh runner cold-starts for you. The full walkthrough, including how to verify where the job ran, is in [Run your first job](/documentation/run-your-first-job). 6. **Optional: invite your team** On every plan except Developer, open **Team** in the dashboard sidebar to invite teammates by email or shareable link (the Developer plan is for individual use, so invites are unavailable there). Invites grant the Admin or Member role. See [Team and roles](/documentation/team-and-roles). ## What to expect in the first hour Here is the typical sequence after you finish onboarding, so you know what is normal and what is worth a second look: > **Nothing showing?** > Make sure a repository is selected in the filter bar at the top of the dashboard, that the repositories you enabled actually ran workflows in the last 30 days, and that the date-range filter overlaps that activity. More in [Troubleshooting](/documentation/troubleshooting). ## Where to go from here - **Learn the dashboard** The layout, every sidebar destination, the filter bar, and the banners that show what Latchkey is doing. - **Move a job to a runner** The one-line change, how pickup works, and how to verify where the job ran. - **Set up notifications** Route pipeline breakage, cost spikes, and trial warnings to email, Slack, or your browser. - **Connect your AI agent** Give Claude Code, Cursor, or any MCP client access to your CI failures, plus opt-in workflow re-runs and one-off jobs on fresh runners. ### How long does Latchkey take to set up? About five minutes: sign in with GitHub, install the app on your organization, and select repositories. The 30-day trial starts on its own, with no separate step. ### How long before I see data in the dashboard? Minutes. A backfill imports the last 30 days of completed workflow runs as soon as you select repositories, so the cost and performance dashboards fill without waiting for new builds. AI Insight findings arrive on the following nightly pass. ### How do I run my first job on a Latchkey runner? Change one line in the workflow: replace the value of `runs-on` with `latchkey-small`. Your steps, actions, secrets, services and matrix stay exactly as they are, and GitHub-hosted labels keep working in the same workflow, so you can migrate one job at a time. --- Latchkey runs CI/CD that repairs its own failures. Agent entry points: https://latchkey.dev/agent.txt, https://latchkey.dev/openapi.json, https://latchkey.dev/llms.txt