Skip to main content
Rippit sits on both sides of the Model Context Protocol, and the two are separate and complementary products.

Connect your MCP client to Rippit

Bring Rippit into your AI assistant. Connect Claude, ChatGPT, Gemini, Microsoft Copilot, or anything else that speaks MCP, and analyze your conversation data from there.

Connect Rippit to your MCP servers

Bring your tools into Rippit. Connect Slack, Intercom, Notion, Salesforce, Snowflake, or any MCP server to Rippit’s own in-app agent.

​
Two directions, one protocol

As an MCP server, Rippit exposes your conversation data to an assistant you already use. Your assistant does the thinking; Rippit supplies the data and the analysis tools. Setup happens in your assistant. As an MCP client, Rippit’s in-app agent connects out to servers you choose. It already has your conversation data, and connecting Slack or Salesforce or Snowflake lets it reach the systems around that data in the same conversation. Setup happens in Rippit, once, by an admin.

​
Most teams use both

These are not competing options, and turning one on has no effect on the other. They solve different halves of the same problem, and the common setup is to run both:
  • The MCP server goes to everyone who already lives in Claude or ChatGPT. It puts conversation data one question away, wherever they were already working.
    • Great for executives and partner teams who live primarily in these tools.
  • Agentic Chat and Agent Apps are where the recurring work lives: the analyses that need your other systems, and the ones that should run whether or not anybody remembers to ask.
Nothing is duplicated between them. The same conversation data, the same permissions, reached two different ways.