Skip to main content

Connecting via MCP (Claude, Cursor, and more)

How to connect Pulse to Claude, Cursor, and other AI tools using the Model Context Protocol.

Written by John Kotowski

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets AI tools connect to external data sources. Pulse supports MCP natively, so you can query live pricing data directly from your AI assistant — every tracked competitor, diff, and digest — without leaving your workflow.

What you can do via MCP

  • Search for companies and their pricing

  • Fetch pricing change history and diffs

  • Manage and read your watchlists

  • Get the latest pricing news

  • Run AI Research skills (Pro)

Supported tools

Pulse connects to any MCP-capable client, including Claude Desktop (recommended), Claude Code, Claude Web, ChatGPT, Codex, Gemini CLI, and others.

Your MCP URL

The Pulse MCP server URL is:

You authenticate with one-click OAuth — there are no API keys to copy or manage for MCP.

How to connect

  1. In the Pulse navigation, click Connect MCP.

  2. Pick the AI tool you use (e.g. Claude Desktop) and click Continue.

  3. Follow the guided, step-by-step instructions shown for that tool.

  4. When prompted, sign in and approve access to authorize the connection.

Example: Claude Desktop

The in-app guide walks you through these steps:

  1. Open Settings → Connectors in Claude.

  2. Choose Add Custom Connector.

  3. Fill out the form, using the MCP URL https://mcp.pricingsaas.com.

  4. Click Connect.

  5. Sign in to Pulse.

  6. Approve access.

Once connected, ask your assistant things like "What are Notion's current pricing tiers?" or "What changed in Squarespace's pricing this month?" and it will query Pulse in real time. The Connect MCP flow also includes a quick "test it" step to confirm the connection works.

Available on all plans

MCP access is included on Free and Pro — no upgrade required to connect. Normal credit costs apply to each tool call; see How credits work. (Note: API and MCP calls are charged per call.)

Prefer the REST API?

If you're building a custom integration rather than connecting an AI tool, see Using the Pulse REST API.

Did this answer your question?