You can wire up the SDK directly, or let Claude / Cursor drive Sheepit for you via MCP. Most vibe-coders do both.
Terminal
npm install @sheepit-ai/sdk-js
your-app.ts
import { Sheepit } from "@sheepit-ai/sdk-js";
const gt = Sheepit.create({
apiKey: "lp_pub_xxx_...", // publishable key from Settings → API Keys
});
// Track your first event
gt.track("app_loaded");
React app? Use @sheepit-ai/react → It wraps the SDK with hooks ( useTrack, useFlag, useExperiment) and a provider.
1 Sign in from the terminal
A browser tab opens, you confirm the project, and a long-lived secret key lands at ~/.sheepit/credentials.json. The MCP server picks it up automatically.
bash
npx @sheepit-ai/cli login
You'll need to create a project first (the button above) so there's something to attach the key to.
2 Install the MCP server in Claude Desktop / Cursor
Auto-detects which MCP-capable clients you have installed and writes thesheepit entry into each one's config file. Restart the client after to see the tools.
bash
npx @sheepit-ai/mcp install --yes
Manual config (if your client isn't auto-detected)
The MCP server exposes ~37 tools — campaigns, dashboards, flags, experiments, insights queries. Try one:
text
Ask Claude:
"Create a dashboard that tracks signups and errors
over the last 30 days for my Sheepit project."
Other prompts that work well: “What are my top traffic sources this week?”, “Create a feature flag called new_pricing_page off by default, then turn it on for users in the dogfooders group.”
3
What success looks like
After your first gt.track() call (or after Claude fires its first tool call), your event shows up live in the Events dashboard — usually within a few seconds.