Pulsiv
Documentation

Pulsiv MCP for AI Agents

Connect Pulsiv to your coding agents and chat assistants so they can list alerts, inspect channels, create alerts, and manage alerts from natural language prompts.

What is an MCP server?

MCP (Model Context Protocol) is a standard way for AI assistants to call external tools safely. Instead of hardcoding custom integrations for each assistant, you expose tools once through an MCP server.

Pulsiv's MCP server gives assistants direct access to your organization's alert workflows so they can create and manage on-chain monitoring tasks for you.

Get a Pulsiv API key

  1. Sign in to Pulsiv and open Settings - API keys.
  2. Create a new key and copy the full secret immediately (it is shown once).
  3. Use that secret as a Bearer token in your MCP client configuration.

Setup in Cursor

Add Pulsiv to your project or global Cursor MCP config:

.cursor/mcp.json
{
  "mcpServers": {
    "pulsiv-mcp": {
      "url": "https://mcp.pulsiv.app",
      "headers": {
        "Authorization": "Bearer pulse_your_org_api_key"
      }
    }
  }
}

Setup in Claude

Run this command in your terminal to register Pulsiv as an MCP server in Claude:

terminal
$ claude mcp add --transport http pulsiv-mcp https://mcp.pulsiv.app \
  --header "Authorization: Bearer pulse_your_org_api_key"

Available Pulsiv tools

ToolWhat it doesType
pulsiv_list_alertsLists your organization's alerts.Read
pulsiv_list_notification_channelsLists notification channels and channel IDs.Read
pulsiv_create_alertCreates a new on-chain alert from contract + event inputs.Write
pulsiv_manage_alertRenames, pauses, resumes, soft-deletes, or hard-deletes an alert.Write