MCP Quickstart - PG:AI Docs

This guide walks you through connecting PG:AI to your AI tool of choice. The whole process takes less than 2 minutes.

Step 1: Generate an API Key

Open PG:AI Settings

Navigate to Settings → API Keys in your PG:AI dashboard

Create a New Key

Click Create API Key and give it a descriptive name (e.g. “Claude MCP” or “Cursor”)

Copy the Key

Copy your API key immediately - it won’t be shown again. Keys start with pgai_live_

Keep your API key secure. Anyone with your key can access your PG:AI account data. If you suspect a key has been compromised, revoke it immediately in Settings.

Step 2: Configure Your AI Tool

Claude.ai
Cursor
Claude Enterprise
Windsurf
Other MCP Clients

Claude supports MCP through Connectors (available on Pro and Enterprise plans).

Open Settings

Go to Settings → Connectors in Claude.ai

Add a Connector

Click Add and select Custom connector

Enter the Server URL

https://mcp.pipelinegeneration.ai/sse

Set Authentication

Choose Bearer Token and paste your PG:AI API key

Save

Click Save - Claude now has access to your PG:AI data

Cursor supports MCP through project-level or global configuration files.Option A: Project-level (recommended for team sharing)Create or edit .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "pgai": {
      "url": "https://mcp.pipelinegeneration.ai/mcp",
      "headers": {
        "Authorization": "Bearer pgai_live_YOUR_KEY_HERE"
      }
    }
  }
}

Option B: Global (available in all projects)Open Cursor Settings → MCP and add a new server with the same configuration.

After adding the configuration, reload Cursor for the connection to take effect. You should see “pgai” listed as a connected MCP server.

Admins can enable PG:AI for their entire organisation.

Open Admin Settings

Go to Admin Settings → Integrations → MCP

Add MCP Server

Add a new MCP server with the following details:

URL: https://mcp.pipelinegeneration.ai/mcp
Authentication: Bearer token with your organisation’s PG:AI API key

Configure Permissions

Choose which users or groups should have access to PG:AI tools

Add PG:AI to your Windsurf MCP configuration:

{
  "mcpServers": {
    "pgai": {
      "serverUrl": "https://mcp.pipelinegeneration.ai/mcp",
      "headers": {
        "Authorization": "Bearer pgai_live_YOUR_KEY_HERE"
      }
    }
  }
}

PG:AI uses standard Streamable HTTP transport. Any MCP-compatible client can connect using:

Server URL: https://mcp.pipelinegeneration.ai/mcp
SSE URL: https://mcp.pipelinegeneration.ai/sse (for clients that require SSE)
Authentication: Authorization: Bearer <your_api_key> header

Refer to your client’s documentation for specific configuration instructions.

Step 3: Test the Connection

Once connected, try asking your AI tool one of these prompts:

Check Companies

“What companies am I tracking in PG:AI?”

Get Insights

“Tell me about [company name] - what are the latest insights?”

Search Contacts

“Search for VP of Engineering contacts in my PG:AI account”

Generate Content

“Generate meeting prep for my call with [company name]”

If your AI tool responds with company data, insights, or generated content - you’re all set.

Troubleshooting

My AI tool says it can't find the PG:AI server

Double-check the server URL. Make sure you’re using https://mcp.pipelinegeneration.ai/mcp (or /sse for SSE-only clients). Restart your AI tool after updating the configuration.

I'm getting authentication errors

Verify your API key starts with pgai_live_ and that it hasn’t been revoked. Make sure the Authorization header uses the Bearer prefix.

The connection works but no data is returned

Your API key may not have the right permissions. Check that it was created with MCP access enabled in Settings → API Keys.

Cursor doesn't show pgai as a connected server

After editing .cursor/mcp.json, reload Cursor (Cmd/Ctrl + Shift + P → “Reload Window”). Check for JSON syntax errors in your configuration file.

Next Steps

Tools Reference

Explore all 5 tools and their parameters

Resources

Learn about passive data resources

Examples

See real-world prompts and use cases

API

Endpoints

MCP Server

​Step 1: Generate an API Key

​Step 2: Configure Your AI Tool

​Step 3: Test the Connection

Check Companies

Get Insights

Search Contacts

Generate Content

​Troubleshooting

​Next Steps

Tools Reference

Resources

Examples

Step 1: Generate an API Key

Step 2: Configure Your AI Tool

Step 3: Test the Connection

Troubleshooting

Next Steps