Competitor ad research, built for AI assistants
Proxy is an MCP (Model Context Protocol) server that gives AI assistants real-time access to four major ad libraries: the Meta Ad Library (Facebook and Instagram), Google Ads Transparency Center, LinkedIn Ad Library, and TikTok EU DSA ad repository.
What we built
Marketers spend hours switching between four separate ad library dashboards to research what competitors are running. Proxy removes that friction entirely. Install it once and ask your AI assistant anything about competitor ads in plain language. Proxy pulls live data from the actual ad libraries and returns it in your chat.
No API keys. No dashboards. No context-switching. Just real-time competitor intelligence inside the AI tools you already use: ChatGPT, Claude, Claude Code, Cursor, and any other MCP-compatible client.
Who we are
Proxy is built by Proxy Intelligence Inc., a Delaware C-Corp. Our founding team comes from Amazon and Canva, where we built products used by millions of people. We started Proxy because we kept running into the same problem: the ad library tools that exist are single-platform, expensive, and completely separate from the AI workflows where strategic decisions actually get made.
How it works
Proxy is built on the Model Context Protocol (MCP), an open standard that lets AI assistants connect to external tools and data sources. When you install Proxy and ask your AI “what ads is Nike running on Google?”, your assistant calls Proxy, Proxy fetches the live ad data from the relevant library, and the results come back directly in your conversation.
There is a free tier with 30 credits per month — no credit card required. Most serious users move to the Max plan at $99 per month, which gives 2,500 credits and 200 per day.
Get started
The fastest way to try Proxy is through ChatGPT: search for “Proxy” in the ChatGPT App Store and connect in minutes. For Claude, Cursor, or any MCP client, run the installer at curl -sL useproxy.dev/install | sh.