Every account becomes legible
Ask what is live, what is winning, what changed, and where the next campaign should point without digging through dashboards.
MCP for growth teams
Put an AI operator inside your growth workflow
Funneled MCP gives your AI tools the account context they need to help your team move from question to campaign direction without the usual platform crawl.
Agent workspace
Growth account context loaded
Find the funnels with the biggest revenue leak this week
Prepare a new post-click page for the founder-story ad angle
Show me reusable blocks from the winning bundle campaigns
Open the funnel that needs QA before launch
Prepared next move
4
journeys to improve
12
winning moves ready
The short answer
MCP (Model Context Protocol) is an open standard that lets AI assistants such as Claude or Cursor use external tools. A funnel MCP server puts a funnel platform behind that standard. The Funneled MCP server gives those assistants tools to build, edit, test, and report on funnels in your Funneled account.
Once connected, an assistant can:
It installs with one command and a one-time sign-in. Connect your AI assistant to the Funneled MCP server.
The dream
The best teams do not need another tab to monitor. They need a way for trusted AI tools to understand the account, gather the facts, and bring back work that is ready for a human decision.
Ask what is live, what is winning, what changed, and where the next campaign should point without digging through dashboards.
Brief an assistant once and let it gather funnel context, product data, analytics, and reusable blocks before the team touches the builder.
Give each client a cleaner operating layer: fewer handoffs, faster audits, better campaign prep, and less repeated platform admin.
How it feels
Ask
Where is revenue leaking? Which journeys deserve more budget? What should the next campaign learn from the last one?
Assemble
Instead of hunting through tabs, your team gets the relevant campaigns, products, performance signals, and reusable work in one prepared view.
Act
Agents help prepare the work, then your team spends its time choosing the direction, refining the story, and getting the campaign live.
Questions
MCP (Model Context Protocol) is an open standard that lets AI assistants such as Claude or Cursor use external tools. A funnel MCP server puts a funnel platform behind that standard. The Funneled MCP server gives those assistants tools to build, edit, test, and report on funnels in your Funneled account: creating funnels and editing drafts, publishing, running A/B tests, and pulling analytics.
Any MCP client that supports stdio servers, including Claude Code, Claude Desktop, Cursor, and custom agents built on the MCP SDK. You add it with the command "npx -y @getfunneled/mcp", then approve access once by signing in to Funneled. No API key is needed.
Approved MCP clients can work with funnels, campaigns, products, media, themes, draft content, permalinks, block libraries, A/B tests, and analytics. Your team can ask an agent to inspect performance, prepare changes, open funnels, and help ship campaign-matched pages.
MCP access is authenticated and scoped to your Funneled account. It is designed for controlled agent workflows where permissions, account context, and product data stay inside the platform boundary.
MCP is for teams that want AI agents to do useful work across their growth stack: agencies managing many clients, operators running frequent campaigns, and technical teams building custom AI workflows.
No. MCP complements the builder. Agents can prepare, inspect, and update work faster, while humans still review the experience, approve changes, and publish what should go live.
Book a demo and see how agents can help your team prepare campaigns, inspect funnels, and reuse winning work inside Funneled.