Model Context Protocol
Your AEO data, insideevery AI tool you use.
Generate one key and connect Claude, ChatGPT, Cursor, Codex or any MCP client to your live PageLens data — visibility, prompts, citations, competitors — so your own agents can answer questions about your brand's AI presence directly.
One key · any MCP client · live data
$ mcp add pagelensfetching manifest … okPAGELENS_MCP_KEY=pk_live_••••
How it connects
From one keyto your first answer.
Three steps, under a minute. No SDK, no glue code.
Generate your key
One click in settings mints a scoped, per-workspace MCP key. It's shown once, tied to your data, and revocable anytime.
Workspace MCP key
shown once · revocable anytime
Add the PageLens server
Drop the server into your client — Claude, Cursor, Codex, ChatGPT desktop — with the standard MCP config. No SDK, no glue code.
[mcp_servers.pagelens]command = "npx"args = ["-y", "@pagelens/mcp"]env = { PAGELENS_MCP_KEY = "pk_live_..." }
Ask your agent about your brand
Your assistant now reads your live PageLens data. Ask it where you're losing citations, which prompts to win next, or how a competitor is trending — and it answers from the source.
“Where am I losing citations this week?”
You dropped on 3 buyer prompts — northwind.com took the top citation on each.
Your visibility on ChatGPT is up 6 points; Gemini is flat.1
Sources
What you get
Your live data,where your agents work.
One key, your whole account
Visibility, prompts, positions, citations, sources, competitors and content — all reachable from a single scoped key.
Works with any MCP client
Claude, ChatGPT desktop, Cursor, Codex, Windsurf — anything that speaks Model Context Protocol.
Always live, never stale
Your agent reads the latest snapshot on every call. No nightly exports, no CSVs to refresh.
Scoped and revocable
Keys are per-workspace and read-scoped. Rotate or revoke in one click; nothing else is exposed.
Build it into your own agents
Wire PageLens into internal assistants and automations so brand-visibility data shows up where your team already works.
No copy-paste, no glue code
Standard MCP config, zero SDK. If your client supports MCP, you're connected in under a minute.
1
key to connect everything
60s
from key to first answer
5+
MCP clients supported
Questions
The details.
- What is MCP / Model Context Protocol?
- MCP is an open standard that lets AI clients connect to external tools and data through a common interface. PageLens runs an MCP server, so any MCP-compatible client can read your PageLens data directly — no custom integration required.
- Which AI clients does it work with?
- Any client that speaks MCP, including Claude, ChatGPT desktop, Cursor, Codex and Windsurf. If your tool supports Model Context Protocol, it can connect to PageLens.
- What data can my agent read?
- Your live PageLens data — brand visibility, tracked prompts and positions, citations and sources, competitors, and content — all reachable from a single scoped key.
- Is it read-only and secure?
- Yes. Keys are per-workspace and read-scoped, so your agent can query your data but nothing else is exposed. You can rotate or revoke a key in one click at any time.
- Do I need to write any code?
- No. You generate a key and add the PageLens server with the standard MCP config your client already uses. There's no SDK to install and no glue code to write.
- Can I use it in my own internal agents?
- Yes. Because it's a standard MCP server, you can wire PageLens into your own assistants and automations so brand-visibility data shows up wherever your team already works.
Put your AEO datawhere your agents work.
Book a demo and we'll connect PageLens to your AI stack in minutes.