MCP connectors: what they are and when to use them

MCP stands for Model Context Protocol — an open standard for letting
LLMs call external tools. We use it to plug your CRM agent into anything
that exposes an MCP server: Meta Ads, Stripe, Linear, your own internal
APIs, third-party services.

Why it matters

Built-in tools cover the common CRM flows — booking, tagging, sending
messages, updating opportunities. MCP covers everything else.

A few examples we've seen:

• Meta Ads — a contact texts about ad performance. The agent fetches
• Stripe — a customer asks about a charge. The agent looks up the
• Linear — a contact reports a bug. The agent files a ticket and
• Your internal API — anything you've already built that you want

How it works

Three pieces:

1. Workspace-level connection — you connect an MCP server once
2. Per-agent attachment — for each agent, you pick which of the
3. Per-tool rules — for each enabled tool, you write a plain-English

The actual tool execution happens via Anthropic's hosted MCP — we pass
the server config through, Claude calls the tool, the result comes
back inline. We log every call so you can audit.

What gets logged

Every MCP tool invocation shows up in the agent's Integrations → Logs
sub-tab as mcp:server-name:tool-name alongside the original
conversation. You can see who triggered it, when, and which tool fired.

Where to start

• Connect an MCP server
• Write rules for when the agent should use a tool

Curated vs custom

We ship one-click cards for Meta Ads, Stripe, and Linear with
sensible default URLs and auth helper text. Anything else? Use the
Custom MCP server option and paste an HTTP MCP URL. Anything that
speaks JSON-RPC over HTTP works.