Your website could be generating enquiries every day, but how many potential customers are slipping through because nobody is following up quickly or qualifying them properly?

A business owner reached out to me with exactly this problem.

Their website was collecting enquiries, but there was no consistent process for reviewing them, identifying serious prospects and responding.

So I built a system to handle that entire process.

Here's what happens when someone submits an enquiry:

→ The system picks it up immediately, whether it's 2 PM or 2 AM.

→ Invalid submissions and duplicate enquiries are filtered out.

→ Each enquiry is assessed against the business's own qualification rules and classified as qualified, needing more information, or not a fit.

→ Every assessment must be supported by exact quotes from the original enquiry. Those quotes are independently verified, so unsupported claims are marked as unknown.

→ Each lead is added to the CRM with its assessment, notes and pipeline card.

→ A personalised reply is drafted based on the information provided, asking only for what's missing.

→ The draft goes to Slack for human review. A team member can approve it, edit it or reject it.

→ Once approved, the email is sent and the lead moves forward in the pipeline.

I built the entire system using n8n and Claude, with the client's existing tools and accounts.

One important detail: AI makes only two decisions throughout the entire workflow. Everything else is designed to validate those decisions and keep the process controlled, traceable and reliable.

The goal isn't simply to automate emails. It's to make sure incoming enquiries are properly handled, qualified and followed up without relying on someone to manually manage every step.

If you're a business owner, founder or agency owner looking to improve how your business handles incoming leads, send me a DM.

Let's build something around your business.

Nice project👍

n8n MCP server: every client you connect sees every workflow you enable If you turn on n8n's instance-level MCP server, the workflows you enable aren't scoped per client. Claude Code, ChatGPT and anything else you connect all see the same list. That's in n8n's MCP server docs: https://docs.n8n.io/connect/connect-to-n8n-mcp-server You enable MCP once, then opt in each workflow. One exception: search_workflows previews every workflow your user can view, without the full data. I use n8n and I write my code in Claude Code. Here's how I'd connect the two without handing over the whole instance: 1. Expose small, single-purpose workflows, like "create draft invoice", never a generic HTTP runner. 2. Leave Auto-expose new workflows off. It's off by default. 3. Use OAuth, grant the narrowest access, and revoke clients you stopped using. 4. Put a human approval step inside any workflow that sends, pays or deletes. The MCP (Model Context Protocol) tools spec says a human SHOULD always be able to deny tool calls: https://modelcontextprotocol.io/specification/2026-07-28/server/tools The tradeoff: one connection serves every client, which is handy until one is a chat app you half trust. For tighter limits, an MCP Server Trigger node exposes tools from just one workflow. Do you give each AI client its own scoped MCP server, or share one?

The docs separate shared workflow visibility from OAuth action permissions. Two clients connected as the same user can see the same enabled workflows, while one is allowed to read and the other to execute.

I'd verify that in staging with a harmless fixture: connect A without...