Atlas Data: Ads, Jobs & Trends

Atlas Data MCP tools

Ad-library, job-market and search-trend data for AI agents, served as MCP tools through the hosted Apify MCP server.

There is nothing to install or host. This repository holds ready-to-paste client configs and the MCP Registry metadata (server.json) for the Atlas Data tool suite.

The Apify MCP server also exposes its generic helpers (get-actor-run, get-dataset-items, get-key-value-store-record, abort-actor-run) so an agent can page through large results.

Endpoint

https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable

• Transport: Streamable HTTP
• Auth: Apify OAuth (clients that support remote OAuth will open a sign-in window) or an Apify API token sent as Authorization: Bearer <token>. Get a token at console.apify.com/settings/integrations.
• Billing: each tool is a pay-per-result Apify Actor, billed to your own Apify account. See each Store page for current pricing.
• Want fewer tools? Remove entries from the tools= list. Each entry is atlasdata/<actor-name>.

Cursor

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project). With OAuth, the URL is enough (configs/cursor-mcp.json):

{
  "mcpServers": {
    "atlas-data": {
      "url": "https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable"
    }
  }
}

To use a token instead of OAuth, add a header and keep the token in your environment (configs/cursor-mcp-token.json):

{
  "mcpServers": {
    "atlas-data": {
      "url": "https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable",
      "headers": {
        "Authorization": "Bearer ${env:APIFY_TOKEN}"
      }
    }
  }
}

Claude Desktop

Option A: custom connector (OAuth). In Claude Desktop open Settings → Connectors → Add custom connector, name it Atlas Data and paste the endpoint URL above. Sign in to Apify when prompted.

Option B: config file with a token. Edit claude_desktop_config.json (Settings → Developer → Edit Config). This uses mcp-remote to bridge the remote server, so Node.js 18+ must be installed (configs/claude_desktop_config.json):

{
  "mcpServers": {
    "atlas-data": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable",
        "--header",
        "Authorization:${AUTH_HEADER}"
      ],
      "env": {
        "AUTH_HEADER": "Bearer <your-apify-token>"
      }
    }
  }
}

Restart Claude Desktop afterwards. (The header is passed through an env var because some clients mangle spaces inside args.)

Claude Code

claude mcp add --transport http atlas-data \
  "https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable" \
  --header "Authorization: Bearer $APIFY_TOKEN"

VS Code (GitHub Copilot agent mode)

.vscode/mcp.json, prompting for the token once (configs/vscode-mcp.json):

{
  "inputs": [
    { "type": "promptString", "id": "apify-token", "description": "Apify API token", "password": true }
  ],
  "servers": {
    "atlas-data": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=atlasdata/google-trends-scraper-reliable,atlasdata/google-ads-transparency-scraper-reliable,atlasdata/linkedin-ad-library-scraper-reliable,atlasdata/tiktok-ad-library-scraper-reliable,atlasdata/google-jobs-scraper-reliable",
      "headers": { "Authorization": "Bearer ${input:apify-token}" }
    }
  }
}

Example prompts

• "Compare Google Trends interest for notion, obsidian and logseq in the US over the last 12 months and list the rising related queries."
• "Which Google ads has hubspot.com run in Germany since September? Group them by format."
• "Pull HubSpot's LinkedIn ads from the last 30 days and summarise the main calls to action."
• "Find active TikTok video ads mentioning protein bar in France and list their landing pages."
• "Get 50 data engineer jobs posted this week in Berlin and give me the salary range where it is listed."

Tool inputs follow each Actor's input schema. The agent sees the schema through MCP; the Store pages and example repos document every field.

Cost control

Every run is billed per result to the Apify account that authenticates. Keep limits such as maxAdsPerSearch and maxResults small while testing, and set a monthly usage limit in the Apify Console if an agent will run unattended.

Calling the same tools from code

The tools are ordinary Apify Actors, so the same data is available without MCP:

from apify_client import ApifyClient

client = ApifyClient("<APIFY_TOKEN>")  # pip install apify-client
run = client.actor("atlasdata/google-jobs-scraper-reliable").call(
    run_input={"queries": ["data engineer"], "location": "Austin, TX", "country": "us", "maxResults": 20}
)
for job in client.dataset(run.default_dataset_id).iterate_items():
    print(job["title"], "|", job["companyName"], "|", job.get("salary"))

More examples (Python, Node.js, curl) are in each tool's repository linked above.

Registry metadata

server.json describes this suite for the official MCP Registry as io.github.AtlasDataDenmark/atlas-data. It is published by the publish workflow using GitHub OIDC.

Support

Open an issue in this repository or in the matching tool repository.

License

The configs and docs in this repository are MIT licensed. The tools themselves are hosted services on Apify.