Connect your AI to your analytics

• "How many visitors this week?"
• "Which pages hold people longest?"
• "Which AI crawlers hit my site, and what did they 404 on?"

It is read only except for one tool that creates a site, and it is scoped to your own data by construction: every call goes through the same site scoped queries the dashboard uses, so a key can only ever read its own team's sites, never raw SQL, and never a visitor's address.

What a key can see. A key belongs to a person, and sites belong to a team, so a key reaches every site of every team you are a member of and no more. Join a second team and its sites appear on the same key. Leave one and they stop. Membership is re-checked on every single call rather than baked into the key, so removing someone closes their access at the same moment it closes their dashboard.

Billing applies here too. The data tools follow the same rule as the dashboard: a team whose subscription is not active drops out of what your key can read, while teams that are still paying are unaffected. Reactivating restores it with no new key and no reconnect.

claude mcp add --transport http peeko https://usepeeko.com/api/mcp
codex mcp add peeko --url https://usepeeko.com/api/mcp
gemini mcp add --transport http peeko https://usepeeko.com/api/mcp

Then ask it: "Use the peeko MCP server to send a Peeko sign in link to my email address."

You finish in a browser. Opening the emailed link is what creates the account, or signs you in, because one link does both. Choosing a size and paying happen on the web. An agent never touches checkout, and this tool never returns a session, a token or a key.

Its answer is the same sentence whether or not the address already has an account. That is deliberate, so nobody can use this endpoint to discover who has a Peeko account. It is rate limited per address and per network, so asking repeatedly will not produce more mail.

Once you are signed up and subscribed, come back for a key. The analytics tools stay invisible until you reconnect with one, and the sign up tool disappears from the list once you do.

The API keys card on your settings page mints one. Name the key, create it, and copy either the raw key or the ready made connect command from the panel that appears once. The same card lists your keys with their prefix, creation date and last use, and revokes them.

Send the key as a bearer token: Authorization: Bearer mcpk_.... It is never a session cookie.

Claude Code

claude mcp add --transport http peeko https://usepeeko.com/api/mcp \
  --header "Authorization: Bearer mcpk_your_key_here"

claude mcp list should then show peeko as connected. The tools load at the start of a new session, or check /mcp in the one you are in.

Codex

codex mcp add peeko --url https://usepeeko.com/api/mcp \
  --header "Authorization: Bearer mcpk_your_key_here"

Claude Desktop, Cursor, or any client with a config file

An HTTP server with a header. Cursor reads ~/.cursor/mcp.json for every project, or .cursor/mcp.json for one.

{
  "mcpServers": {
    "peeko": {
      "type": "http",
      "url": "https://usepeeko.com/api/mcp",
      "headers": { "Authorization": "Bearer mcpk_your_key_here" }
    }
  }
}

Then ask the assistant to list your sites and go from there. Call the explain_metrics prompt for exact metric definitions, including the one that catches people out: visitors are distinct daily hashes, so a multi day total sums daily uniques.

Ranges are today, 24h, 7d, 30d, 90d and all, defaulting to 7d. Filters are any of page, source, country, device, browser, browser_version, os, os_version, screen, viewport, utm_source, utm_medium and utm_campaign, and they are the same filter model as the dashboard, narrowing every metric at once.

Plus one prompt, explain_metrics, defining every metric and the privacy model.

Creating a site from your agent

create_site takes a name and a domain and hands back the site id, the public site key, and a prompt with that real key already in it, ready to paste into a coding agent. It seeds the site's allowed origins with your domain and its www version, so a correctly installed snippet records traffic straight away.

If your key reaches more than one organization you must pass org. Peeko will not pick one for you: putting a site in the wrong organization bills it to the wrong party and shows it to the wrong people, and nothing would ever report it.

404s are not observable on every site

The crawler middleware runs before your response exists, so it can only report a response status where your framework exposes one after the fact. Some frameworks have such a hook. Next's not-found boundary is not one, because it also renders for requests that resolve 200.

A crawler's identity is a claim, not a fact

Classification is done from the user agent alone, and Peeko stores no address on the crawler path by design. Anything can send Googlebot's user agent, and it will be reported as Google. Treat crawler names as self reported.

• Transport. Streamable HTTP, specification 2025-06-18, hand rolled. Post a JSON-RPC request and get one application/json response. The server never initiates messages, so there is no server sent event stream and no session state, and GET and DELETE return 405.
• Auth. A static bearer API key, with no OAuth flow. A missing or invalid key gets a 401.
• Errors. JSON-RPC error codes for protocol problems, -32601 for an unknown method and -32602 for bad arguments or an unknown tool. A domain failure, such as a site you do not own, comes back as a tool result marked as an error rather than as a protocol error.

Locally the base address is http://localhost:3000 instead of the production origin. Everything else is identical.

• Put Peeko on your site is what you send somebody. It covers Lovable, v0, Bolt, Cursor, Claude Code, Codex, Next.js and plain HTML, and repeats the client configs above.
• The Peeko command line is the second front door onto this same endpoint, for when you would rather not leave the terminal.