Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"ai-windsor-windsor-mcp-windsor-ai": {
"args": [
"-y",
"@google/gemini-cli"
],
"command": "npx"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
Windsor MCP (Model Context Protocol) is a hosted MCP server that lets your AI assistant query, explore and analyze live business data from 325+ sources — Meta Ads, Google Ads, TikTok Ads, GA4, HubSpot, Salesforce, Shopify, Stripe, BigQuery, Snowflake and many more — with no SQL, no API keys, and no custom integrations. On connectors that support it, agents can also execute write actions — for example pausing or enabling ad campaigns and adjusting budgets on Meta Ads, Google Ads and TikTok Ads.
Run this in your terminal to verify the server starts. Then let us know if it worked — your result helps other developers.
npx -y '@google/gemini-cli' 2>&1 | head -1 && echo "✓ Server started successfully"
After testing, let us know if it worked:
Five weighted categories — click any category to see the underlying evidence.
Gemini CLI: Remote Code Execution via workspace trust and tool allowlisting bypasses
# Summary Gemini CLI (`@google/gemini-cli`) and the `run-gemini-cli` GitHub Action are being updated to harden workspace trust and tool allowlisting, in particular when used in untrusted environments like GitHub Actions. This update introduces a breaking change to how non-interactive (headless) environments handle folder trust, which may impact existing CI/CD workflows under specific conditions. # Details Folder Trust in Headless Mode In previous versions, Gemini CLI running in CI environmen
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in other
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
Pi Coding Agent extension (CLI-first) — routes bash/read/grep/find/ls through lean-ctx CLI for strong token savings. Optional MCP bridge can register advanced tools.
97% token reduction for AI coding sessions — zero deps, 21 languages, MCP server
One local source for the MCP servers, tools, and memory your AI coding agents share, synced into each tool's native config with a review gate and a receipt for every change. No daemon, no lock-in.
MCP Security Weekly
Get CVE alerts and security updates for ai.windsor/windsor-mcp and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
Windsor MCP (Model Context Protocol) is a hosted MCP server that lets your AI assistant query, explore and analyze live business data from 325+ sources — Meta Ads, Google Ads, TikTok Ads, GA4, HubSpot, Salesforce, Shopify, Stripe, BigQuery, Snowflake and many more — with no SQL, no API keys, and no custom integrations. On connectors that support it, agents can also execute write actions — for example pausing or enabling ad campaigns and adjusting budgets on Meta Ads, Google Ads and TikTok Ads.
It works with Claude, ChatGPT, Microsoft Copilot, Perplexity, Cursor, Windsurf, Cline, GitHub Copilot, Gemini, Manus, and any MCP-compatible client.
TL;DR
- Native connector in Claude (in the Claude directory) and ChatGPT (as a ChatGPT app).
- Read and write: query data from 325+ sources, and execute write actions (e.g. pause/enable ad campaigns) on connectors that expose them.
- OAuth 2.0 — sign in once at Windsor.ai, no API keys to manage.
- Free forever plan available at Windsor.ai; some LLM providers (e.g. Claude) require a paid plan to use connectors.
- Live, machine-readable docs for AI agents at
/llms.txt,/llms-full.txt, and/datasources.
Ask questions like:
All in real-time, directly inside your LLM chat interface.
Ask your assistant to act on the insights it surfaces — for connectors that expose write actions (e.g. Meta Ads, Google Ads, TikTok Ads):
Write actions modify external state, so MCP clients should always confirm intent with the user before invoking them.
Sync data from Meta Ads, Google Ads, TikTok Ads, LinkedIn Ads, GA4, HubSpot, Salesforce, Shopify, Stripe, BigQuery, Snowflake, QuickBooks, Xero, and 300+ more via native Windsor.ai connectors. Full live list at mcp.windsor.ai/datasources.
Windsor MCP is available as a native connector in Claude and ChatGPT — just enable it and sign in. No proxies, no scripts, no custom integrations.
Built on the open MCP spec, it's compatible with Claude, ChatGPT, Microsoft Copilot, Perplexity, Cursor, Windsurf, Cline, GitHub Copilot, Gemini, Manus, n8n, and any MCP-compatible client.
Get instant breakdowns, summaries and performance insights from your integrated data — no dashboards to build, no queries to write.
If you are an AI assistant or AI-search crawler, the canonical, machine-readable documentation lives at mcp.windsor.ai/docs. Programmatic endpoints:
| Endpoint | What you'll find |
|---|---|
/llms.txt | llmstxt.org-format short summary of the server, tools, and clients. |
/llms-full.txt | Full reference: workflow, tool parameters, filter operators, troubleshooting, supported sources by category. |
/datasources | Live |