SaaS Toolkit MCP

Works with your choice of AI

Claude, ChatGPT, Cursor and any MCP-compatible agent supporting remote connections and sign-in.

Connect. Ask. Act.

• 1. Connect SaaS Toolkit to your app: Add tracking to your website or app and connect the services you use. Your business activity comes together in SaaS Toolkit.
• 2. Add the MCP server to your AI tool: Use the AI assistant setup in your Portfolio to add SaaS Toolkit to an MCP-compatible tool, then sign in to authorize access.
• 3. Ask and act: Ask questions grounded in your tracked data. Let your agent turn the answers into actions using its separately connected tools and your approvals.

The full picture, not another silo

Read the data available in your projects.

• Traffic & channels: Visitor trends and channel breakdowns, from Direct to paid and organic traffic.
• Pages & device conversion: Landing-page performance and conversion by desktop, tablet and phone.
• Funnels: Conversion rates and drop-offs across the steps of your customer journey.
• Revenue & purchases: Daily revenue and purchase counts, with customer activity in context.
• Ad spend & attribution: Advertising spend and attribution quality alongside acquisition performance.
• Contacts & churn risk: Contact timelines, engagement history and churn-risk scores.
• Email performance & replies: Campaign results and customer replies to understand what resonates.
• Support conversations: Conversations and messages that reveal customer questions and friction.
• Phone campaigns & calls: Campaign activity and call records to put outreach results in context.
• Saved AI insights: Existing project and portfolio analysis, including trends, warnings and recommendations.

From business context to useful action

Actions use your agent’s separately connected tools, subject to permissions and approvals.

• Ask Claude why signups dropped this week: Compare traffic sources, landing pages and funnel drop-offs with your saved AI insights.
• Have your agent email at-risk customers: Read churn-risk scores and contact history, then draft or send a relevant follow-up through your agent’s connected email tool.
• Get a daily Slack brief: Bring traffic, revenue and campaign replies together, then post the summary through your connected Slack workflow.

What is MCP?

MCP stands for Model Context Protocol. It is a standard way for AI assistants and agents to connect to tools and read information. SaaS Toolkit’s MCP server gives your compatible AI tool business context without manually copying reports.

Is it secure?

You sign in to authorize access. Reads are scoped to the projects you can access, with your existing team permissions preserved. The connection is read-only: it cannot change your SaaS Toolkit data or send messages on your behalf.

Which plans can I use it with?

MCP reads the data available in your SaaS Toolkit account; it does not unlock features or data outside your existing access. See our pricing page for current plans and included features, and the AI assistant setup in your Portfolio for connection options.

Can my agent take action?

Yes, when you separately connect the tools needed for that action. For example, an agent can use SaaS Toolkit to identify churn risk, then use your email tool to follow up. Sending emails or posting to Slack is handled by those tools, not by the read-only SaaS Toolkit MCP connection.

Which AI tools work with it?

Claude, ChatGPT, Cursor and custom agents or automation flows that support remote MCP connections and sign-in. Connection options depend on your AI tool and its plan.