# PumpGTM MCP server

Connect an AI agent to PumpGTM to find interested buyers and run outreach through LinkedIn, email, and X. Replies go to a human for review.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 4 repo stars
- Author: pumpgtm
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/pumpgtm/pumpgtm-mcp
- Updated: 2026-10-19
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/pumpgtm-mcp-server

## What it does

PumpGTM gives an agent access to a workspace where it can look for buyers showing intent and start outreach sequences across LinkedIn, email, and X. When someone replies, the response is handed to a person to decide what happens next. That keeps the agent involved in outreach while leaving replies under human control.

The hosted server uses Streamable HTTP at `https://mcp.pumpgtm.com/mcp`. Sign in with OAuth 2.1 or use a workspace key; each credential is scoped to one workspace. In Claude Code, add the server with an MCP command, then authenticate. The agent cannot access the workspace until you sign in.

## Requirements
- Remote server: `https://mcp.pumpgtm.com/mcp`
- Works with: LinkedIn

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http mcp https://mcp.pumpgtm.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add pumpgtm-mcp-server --url https://mcp.pumpgtm.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "pumpgtm-mcp-server": {
            "url": "https://mcp.pumpgtm.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "pumpgtm-mcp-server": {
            "type": "http",
            "url": "https://mcp.pumpgtm.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http pumpgtm-mcp-server https://mcp.pumpgtm.com/mcp
```

## When to use it
- find buyers showing intent
- start an outreach sequence on LinkedIn, email, or X
- route incoming replies to a human
- connect an AI agent to a PumpGTM workspace

## Example prompts
- Find buyers showing intent and add them to my PumpGTM workspace.
- Start an outreach sequence for these buyers on LinkedIn, and send their replies to me.
- Connect to PumpGTM and find people to contact through X.

## Similar
- [outreach-specialist](https://affpapa.org/en/nevibecode/skills/outreach-specialist.md) — skill · 1400 installs
- [Oneprofile](https://affpapa.org/en/nevibecode/mcp/oneprofile.md) — MCP server
- [cold-outreach](https://affpapa.org/en/nevibecode/skills/cold-outreach.md) — skill · 544 installs
- [Omentir](https://affpapa.org/en/nevibecode/mcp/omentir.md) — MCP server · ★ 56 repo stars
- [HeyLead — AI agent for LinkedIn outreach](https://affpapa.org/en/nevibecode/mcp/heylead-ai-agent-for-linkedin-outreach.md) — MCP server · ★ 13 repo stars
- [LinkMCP: hosted LinkedIn MCP server](https://affpapa.org/en/nevibecode/mcp/linkmcp-hosted-linkedin-mcp-server.md) — MCP server · ★ 4 repo stars

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
