Mimetic — Website Growth and Analytics MCP server

mim: the Mimetic CLI

Bring your site's growth context into the terminal and the AI tools where you already work.

Quick start

Install it, then run setup. Setup handles sign-in, project detection, and MCP registration for whichever agents this machine has, and verifies each step for real rather than assuming it worked.

npm install -g @mimeticinc/mim-cli
mim setup

Setup registers the server with Claude Code, Codex, Cursor, Windsurf and Claude Desktop, whichever of them it finds. Each config file is backed up before it is written, and any other MCP servers you have are left alone. Add one later with mim mcp install cursor, or re-run mim mcp install all.

Install globally rather than through npx. Your agent launches mim as the MCP server, so it needs a mim that stays on your PATH, and every example in these docs is written as mim . Requires Node.js 20 or newer.

Re-running setup is safe: it reuses your existing login and registration instead of creating new credentials.

If mim says "command not found" after the install, npm's global bin directory is not on your shell PATH. Run this to see the exact fix for your shell:

npx -y @mimeticinc/mim-cli setup

That prints the export PATH= line to add to your shell profile. Use the npx form for one-off commands too, but install globally before wiring up an MCP client.

Mimetic connects audits, live analytics, ad performance, lifecycle marketing, and session behavior so you can find what is holding growth back, decide what to do next, and move into a reviewable fix workflow without hopping between dashboards. Use mim directly from the shell or give the same context and actions to Claude Code, Codex, and other MCP clients.

Mimetic is the hosted service behind the CLI. Learn more at trymimetic.com.

What you can accomplish

and redacted session-replay insights into one project context.

• Find the highest-impact problems. Pull site audits, prioritized findings,

revenue, search, advertising, product analytics, and lifecycle performance without exporting CSV files.

• Answer growth questions with live data. Query acquisition, conversion,

Console, install tracking through a PR, Shopify pixel, or agent-applied snippet, and create or connect Google Ads resources.

• Repair measurement and acquisition setup. Provision hosted GA4 and Search

the proposed PR, diff, preview, or Mimetic review page.

• Turn findings into reviewable work. Start a code-fix workflow, then open

findings and connected data before it recommends or starts an action.

• Give coding agents business context. Let an MCP client inspect the same

Integrations

Shortened. The full README is on GitHub.