Connect AI Agents to Your Ad Server

What is the AdGlare MCP server?

The Model Context Protocol is an open standard that lets AI applications connect to external tools and data. The AdGlare MCP server exposes your ad server as a set of tools an AI agent can discover and call. For example, you can ask:

• “Which of my campaigns are active right now?”
• “Show impressions, clicks and CTR for campaign 12 last week.”
• “Summarize yesterday’s performance and flag anything unusual.”

It works with any MCP-compatible client, so you do not need a custom integration for each AI system.

Quick start

1. Get your API key. Log in to AdGlare and open your account Settings to find your API key.
2. Find your MCP URL. Every customer has a dedicated endpoint:

   https://{your_name}.mcp.adglare.app/mcp

   Replace {your_name} with your AdGlare account name.
3. Add the server to your AI agent using one of the examples below, with your key sent as a Bearer token.
4. Ask a question, for instance “List my active campaigns.”

Authentication with API keys

Every request must include your API key in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Requests without a valid key receive an HTTP 401 response. Never paste your key into a chat message or commit it to a code repository.

Available tools

AI agents discover these tools automatically when they connect.

Set up the MCP server in your AI agent

Replace {your_name} and YOUR_API_KEY in each example. Client configuration formats change over time, so check your client’s own MCP documentation if something differs.

Claude Code

claude mcp add --transport http adglare https://{your_name}.mcp.adglare.app/mcp \
  --header 'Authorization: Bearer YOUR_API_KEY'

Claude Desktop

Add this to claude_desktop_config.json. It uses the mcp-remote bridge, which requires Node.js.

{
  'mcpServers': {
    'adglare': {
      'command': 'npx',
      'args': ['-y', 'mcp-remote', 'https://{your_name}.mcp.adglare.app/mcp',
               '--header', 'Authorization:'],
      'env': { 'ADGLARE_AUTH': 'Bearer YOUR_API_KEY' }
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project, or the global Cursor config.

{
  'mcpServers': {
    'adglare': {
      'url': 'https://{your_name}.mcp.adglare.app/mcp',
      'headers': { 'Authorization': 'Bearer YOUR_API_KEY' }
    }
  }
}

VS Code (GitHub Copilot)

Add to .vscode/mcp.json. VS Code asks for the key once and keeps it out of the file.

{
  'servers': {
    'adglare': {
      'type': 'http',
      'url': 'https://{your_name}.mcp.adglare.app/mcp',
      'headers': { 'Authorization': 'Bearer ${input:adglare-key}' }
    }
  },
  'inputs': [
    { 'type': 'promptString', 'id': 'adglare-key', 'description': 'AdGlare API key', 'password': true }
  ]
}

Gemini CLI

Add to ~/.gemini/settings.json.

{
  'mcpServers': {
    'adglare': {
      'httpUrl': 'https://{your_name}.mcp.adglare.app/mcp',
      'headers': { 'Authorization': 'Bearer YOUR_API_KEY' }
    }
  }
}

OpenAI API (Responses API)

from openai import OpenAI

client = OpenAI()
response = client.responses.create(
    model='YOUR_MODEL',
    tools=[{
        'type': 'mcp',
        'server_label': 'adglare',
        'server_url': 'https://{your_name}.mcp.adglare.app/mcp',
        'headers': {'Authorization': 'Bearer YOUR_API_KEY'},
        'require_approval': 'always',
    }],
    input='Which AdGlare campaigns are active?',
)
print(response.output_text)

Test your connection

This request should return the list of tools above:

curl -s https://{your_name}.mcp.adglare.app/mcp \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{'jsonrpc':'2.0','id':1,'method':'tools/list'}'

Security and best practices

• An agent can only access what your API key can access. Treat the key like a password.
• Store keys in environment variables or your client’s secret storage, not in shared files.
• Revoke or replace a key in your AdGlare settings if it may have been exposed.
• Keep tool approval turned on in your AI client so you can review actions before they run.
• Data returned by tools is sent to the AI provider you use. For maximum control, use a locally hosted model.

See also the AdGlare security and privacy pages.