Skip to content
Agent Analytics

Cursor

Cursor

For Cursor, start with the Agent Analytics skill plus CLI-style workflows. In practice this is usually faster, cheaper on tokens, and easier to steer than MCP for day-to-day use.

If you are doing this from a Paperclip company, use Set up Agent Analytics for your Paperclip company for the company-skill path. Use this page when you need the raw Cursor install flow itself.

Prerequisites

Section titled “Prerequisites”
  • Cursor installed
  • npx available in the environment Cursor uses
  • Access to the Google or GitHub sign-in you want to connect when Cursor opens browser approval
Section titled “Recommended: install the agent skill”
Terminal window
npx skills add Agent-Analytics/agent-analytics-skill

If the installer asks which skill to install, choose agent-analytics for setup and reporting. Choose agent-analytics-autoresearch when you want Cursor to run a structured growth loop for experiment candidates.

Then ask Cursor:

Set up Agent Analytics for this project. Install it here if needed. Open the browser for me or give me a login link, then wait. I will sign in with Google or GitHub, approve it, and paste back any finish code if you need it. Then create the project, add tracking and key events, and verify the first event.

This gives Cursor the Agent Analytics workflow layer plus CLI-oriented execution in the same environment. That is usually the best tradeoff when you want lower latency and less token overhead than MCP tool calls. The browser approval step creates or connects the account during setup, so you do not need to prepare an API key first.

Verify the install

Section titled “Verify the install”

Ask Cursor:

  • List my Agent Analytics projects
  • How is my-site doing this week?
  • What are the top pages for my-site this week?

If you have not created your first real project yet, continue with First Project in 5 Minutes.

Alternative: add a custom MCP server

Section titled “Alternative: add a custom MCP server”

Use MCP in Cursor if you specifically want connector-style tool calls instead of skill + CLI execution.

  1. Open the Command Palette
  2. Search for MCP
  3. Choose Cursor Settings > Tools & MCP
  4. Click Add Custom MCP
  5. Add this to your mcp.json:
{
  "mcpServers": {
    "agent-analytics": {
      "url": "https://mcp.agentanalytics.sh/mcp"
    }
  }
}

Save the file and reload Cursor if the tool list does not refresh automatically. MCP works, but it usually adds more latency and token overhead than the skill path.

Fallback: use the CLI directly

Section titled “Fallback: use the CLI directly”

If you do not want the skill layer, fall back to the official CLI before you drop to raw HTTP:

Terminal window
npx @agent-analytics/cli@0.5.16 login
npx @agent-analytics/cli@0.5.16 projects
npx @agent-analytics/cli@0.5.16 stats my-site --days 7

That keeps Cursor on the same shell-first path without switching to manual auth headers and raw request payloads. For login behavior, command coverage, and CLI-to-API mapping, use the dedicated CLI page.

Troubleshooting

Section titled “Troubleshooting”
  • If the skill installs but the setup pauses on approval, complete the browser sign-in and paste back any finish code if Cursor asks for it.
  • If you intentionally use the advanced/manual API-key path, confirm AGENT_ANALYTICS_API_KEY is available in the environment Cursor actually uses.
  • Confirm the mcp.json entry is valid JSON if you choose the MCP path.
  • Reload Cursor after adding the custom MCP server if the tools panel still shows the old state.
  • If Cursor can see the MCP server but not your projects, verify the hosted sign-in completed with the correct account.
Section titled “Related”