MCP Server

Connect Pulsify to your LLM agent via the Model Context Protocol. Query your listings, events, and automations through natural language. Available on the Growth plan.

What is MCP?

The Model Context Protocol (MCP) is a standard for connecting AI assistants to external tools and data. Pulsify's MCP server lets your LLM agent access your Amazon data directly, without writing code.

Setup

Add Pulsify to your MCP client configuration. You'll need your API key from the account settings.

{ "mcpServers": { "pulsify": { "command": "npx", "args": ["-y", "@pulsify/mcp-server"], "env": { "PULSIFY_API_KEY": "your-api-key" } } } }    

Available tools

The MCP server exposes the following tools to your LLM agent:

• list_listings — get all your monitored listings with current prices
• get_listing — get details for a specific ASIN or SKU
• list_events — get recent events for a listing
• get_automation — get the current automation code for an event type
• update_price — change a listing's price (respects guardrails)
• update_guardrails — change floor and ceiling for a listing

Example prompts

Once connected, you can ask your LLM agent to query data:

• "What's my current price on ASIN B08N5WRWNW?"
• "Show me the last 10 events for SKU WIDGET-001"
• "Which listings lost the Buy Box today?"

Or ask it to write automation code for you:

• "Write code that matches the Buy Box price when I'm not winning"
• "Create an automation that pauses my campaign when I lose the Buy Box"
• "Write code that sends a Slack alert when a competitor undercuts me by more than 10%"
• "Help me write a strategy that nudges my price up by 1 cent when I'm winning"

Compatible clients

The Pulsify MCP server works with any MCP-compatible client, including Claude Desktop, Cursor, and custom agents built with the MCP SDK.