What it means
The Model Context Protocol (MCP) is an open standard that lets an AI model connect to outside tools and data through one common interface. Anthropic released it in November 2024 as a standard for connecting AI assistants to the systems where data lives. A tool publishes an MCP server, and any AI client that speaks MCP can use it.
For GTM teams it means Claude can read a CRM record, check a campaign or update a task without anyone pasting data into the chat. Each server exposes its own set of actions, so what Claude can do depends on what the tool's server offers and the access you grant it. A tool without a server is still reachable through its API.
How we use it
MCP is how our Claude setup reaches the live stack. Workspace tools such as Slack, ClickUp and Google Drive connect through MCP connectors, and a shared skill library reaches HeyReach, Attio, Supabase and BounceBan the same way. Sending tools such as Smartlead and EmailBison are called through their APIs from our skills.
Setting up connectors, including custom ones for tools outside the directory, is covered in our Claude GTM playbook for beginners. How the connections fit the rest of the system is in Claude Code for GTM.