Rémi KokabiNovember 4, 20266 min read

Webflow integration with lemlist through MCP, with prompts to copy

A Tier 1 account visits your pricing page twice, downloads a case study, then goes quiet. That behavior lives in Webflow. Your follow-up sequence lives in lemlist. Nobody connects the two, so the account gets the same generic email as a lead who showed zero interest.

A real webflow lemlist integration through MCP closes that gap. Connect both MCP servers to the same AI client and an agent can read what’s happening on your site, then act on it directly inside your campaigns.

In this article, I’ll show you exactly how that works, with prompts you can copy into Claude, ChatGPT or Cursor today. You’ll see how to turn a Webflow page into a trigger for personalized outreach, and what guardrails to keep in place before you let an agent write in both tools at once.

Why connect lemlist and Webflow through MCP

A page built for one account shows effort a template never can. When both the lemlist MCP and the Webflow MCP are connected in the same AI client, an agent can create a CMS page per account in Webflow and drop its URL straight into the matching lemlist sequence, all from one conversation.

An MCP server is a standardized layer that lets an AI agent read and write data in an external tool through natural language. Instead of you copying data between tabs, the agent calls the tool’s API on your behalf and returns structured results. You can learn more about how this works in practice in this guide to MCP for sales.

That combination turns Webflow from a marketing site into an outbound asset. Every personalized landing page the agent publishes becomes a link your prospect sees in their inbox, tied to the exact campaign step where it matters most.

What you need before you start

Three things should be in place before you prompt anything:

  • The lemlist MCP connected in Claude, ChatGPT or Cursor. Follow the walkthrough on how to set up the lemlist MCP if you haven’t done this yet.

  • The Webflow MCP connected in the same AI client, authorized against the Webflow site you want the agent to edit.

  • Permissions in both tools scoped to what the use cases below actually need. If the agent only needs to read CMS items, don’t grant it full site admin access.

Start with read-only prompts. Ask the agent to list your Webflow collections or pull stats from a lemlist campaign. Once you trust the output, move to prompts that write data.

1. One page per Tier 1 account

Your Tier 1 accounts get the same generic link as every other lead in the sequence. The page they land on says nothing about their company, their industry, or the problem you solve for people like them.

Here is the prompt to copy:

Create a CMS page for each of my 10 Tier 1 accounts in Webflow, with their name, industry and one relevant case study, and add each URL as a variable in step 2 of campaign X.

The agent reads your Webflow CMS collection, creates one page per account with the fields you specified, publishes each page, then writes the live URL into a custom lead variable inside the lemlist campaign. When the sequence reaches step 2, each lead sees a link to a page that was built for their company.

Before running this on a full list, open one of the published pages and confirm the layout, the case study match, and the URL format. A wrong slug or a broken layout at scale is harder to fix than catching it on the first page.

This approach fits naturally into an account based prospecting workflow where every touchpoint is tailored to the account, not just the person.

2. A calculator for a campaign

Telling a RevOps leader you can save them hours every week is a claim. Letting them type their own numbers into a calculator and see the result is proof they generate themselves.

Here is the prompt to copy:

Build a simple ROI calculator in Webflow for our sequence to RevOps leaders and add its link to the last step.

The agent creates a new page in Webflow with input fields and a basic calculation logic, publishes it, then updates the final step of your lemlist campaign to include the calculator URL. The prospect receives the link at the end of the sequence, after they already have context from earlier steps.

Before launching, open the calculator page, enter realistic numbers, and verify the output makes sense. A calculator that returns a wrong figure will hurt credibility more than no calculator at all.

Guardrails when an agent works in lemlist and Webflow

Giving an agent write access to two tools at once means you need clear boundaries. These guardrails keep things under control:

  • Review every campaign before launch. The agent can create and update campaigns, but a human should approve the final version before any lead receives a message.

  • Test on 10 leads first. Run the full workflow on a small batch, check the pages and the sequence, then expand.

  • Keep unsubscribes and exclusions on. Never let a prompt override your suppression lists or compliance settings in lemlist.

  • Decide who on the team can connect the MCPs. One person with admin access creating a connection is different from every rep running their own agent with full permissions.

  • Log what the agent changed. After each session, ask the agent to list every page it created in Webflow and every campaign or lead it modified in lemlist. Save that log.

Run it in lemlist

The lemlist MCP server works in Claude, ChatGPT and Cursor. Once connected, the agent can do more than just update campaign steps.

It can search the People database with 650M+ contacts and 65M+ companies, then enrich the results with waterfall enrichment across 14+ providers to get verified emails and phone numbers. From there, it can create campaigns and sequences across email, LinkedIn, calls and WhatsApp, read the unified inbox and campaign stats, and act on intent signals to prioritize leads who are showing buying behavior right now.

That means the same conversation where you build Webflow pages can also source the leads who will receive them, enrich their contact data, and launch the sequence. You can explore the full list of lemlist MCP use cases to see what else the agent handles.

Final takeaway

Webflow and lemlist don’t share a native integration, but they don’t need one. With both MCP servers connected in the same AI client, an agent builds account specific pages in Webflow and wires them into lemlist sequences in a single prompt. The prompts above are ready to copy, and the guardrails keep you in control while the agent does the manual work.

Build your first signal-based sequence in lemlist.

Launch AI outbound campaigns

Automate prospecting and personalize AI-powered sequences to book more meetings.

Frequently asked questions

Through MCP, yes. Connect both MCP servers in the same AI client and the agent can read Webflow and act in lemlist in one conversation.

Any client that supports remote MCP servers, such as Claude, ChatGPT or Cursor, with both servers connected.

It can create and update campaigns, but review each campaign before launch and test on a small batch first.

No. You connect both MCP servers in your AI client, then write your requests in plain language.
Rémi Kokabi

Rémi Kokabi

Co-founder @GTM Club

Share this post