AI workflow MCP-native Claude · ChatGPT · Gemini · MCP
The Pipeline Engine
You describe the kind of client you want to work with, in your own words. The Pipeline Engine finds matching companies, identifies the right decision-maker at each, verifies every email, pulls one fresh public signal per lead (a recent project, a launch, a new hire, a press mention), and writes you a tailored opening line you can drop into a cold email or DM. Output to CSV, Google Sheets, or Notion. You write the actual message. Your taste is why it lands.
What it does
- Runs on Claude, ChatGPT, or Gemini.
- Pulls candidate companies by your ICP, in any region you specify.
- Finds the right decision-maker at each company.
- Verifies every email before it reaches you.
- Researches each lead and writes you a one-line opener.
- Delivers results as a CSV, a Google Sheet, or a Notion database.
How it works
- Buy the skill. You get a single bundle, ready to paste into Claude, ChatGPT, or Cursor. No code to write or maintain.
- Connect three free-tier services to your AI tool, once. Apify for discovery and research. Hunter or Snov.io for email verification. CSV, Google Sheets, or Notion for output. About ten minutes.
- Open a new chat and say "run the Pipeline Engine". Describe the kind of client you want to find. Pick your output. Confirm the spend before anything runs.
- It does a dry-run of five leads first so you can check the quality before any bigger pull. Approve and it goes for the full list.
Install
Our AI workflows are designed to be fully platform agnostic - meaning they are compatible with your AI platform of choice.
01 Claude Code Local install · recommended
Drop the unzipped folder into your Claude Code skills directory and restart.
- Unzip the pipeline-engine bundle from your purchase email.
- Copy the whole pipeline-engine/ folder into ~/.claude/skills/.
- Add the Apify MCP snippet below to ~/.claude/mcp.json. Restart Claude Code.
- In any project, say "run the Pipeline Engine".
{
"mcpServers": {
"apify": {
"url": "https://mcp.apify.com",
"headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
}
}
}02 Claude Desktop MCP-native
Paste SKILL.md into a Project and connect the Apify MCP. The skill is live in every chat in that Project.
- Open Claude Desktop → Settings → Developer → Edit Config. Add the snippet below under mcpServers.
- Restart Claude Desktop.
- Create a Project named "Pipeline Engine". Paste SKILL.md from the bundle into Project Instructions.
- Open a chat in the Project and say "run the Pipeline Engine".
{
"mcpServers": {
"apify": {
"url": "https://mcp.apify.com",
"headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
}
}
}03 ChatGPT Project or Custom GPT
Build a Project or Custom GPT that wraps SKILL.md. Requires a plan with MCP connectors.
- Open ChatGPT → Settings → Connectors → Add MCP server. Paste the Apify MCP URL.
- Create a new Project (or Custom GPT). Paste SKILL.md into Instructions.
- Open a chat with the Project or GPT and say "run the Pipeline Engine".
04 Cursor MCP-native
Drop the bundle into a folder Cursor reads and reference SKILL.md from your chat.
- Copy the pipeline-engine/ folder into ~/.cursor/skills/ (or any folder Cursor watches).
- Open Cursor Settings → MCP → Add Server. Paste the Apify MCP URL.
- In a chat, reference @pipeline-engine/SKILL.md and say "run the Pipeline Engine".
Stacks with
These AI workflows are designed to chain together. Run one's output through another and you get a real production pipeline.
- AI Proposal Drafter
Pairs with AI Proposal Drafter from MWM Issue 03. When a Pipeline Engine lead books a discovery call, hand the transcript to AI Proposal Drafter and the SOW comes back in your studio's voice.
- AI Editor In Chief
Pairs with AI Editor In Chief from MWM Issue 02. The Pipeline Engine writes the opener. AI Editor In Chief writes the rest of the email through your voice fingerprint.