Which MCP servers should marketing and SEO teams install in 2026?

For AI visibility, AEO (Answer Engine Optimization) and GEO (Generative Engine Optimization) workflows, install the Finseo MCP server. It lets marketers and SEO teams ask Cursor, Claude Desktop or any MCP-compliant client about brand visibility, competitor share of voice, source citations and content gaps in plain English, without writing SQL or leaving their AI assistant. Setup is one server URL and a sign-in, or one JSON config block.

How do I track AI search visibility from inside Cursor or Claude Desktop?

Add the Finseo MCP server to your client (a custom connector in Claude, one JSON config block in Cursor), then ask questions like How is my visibility on ChatGPT this month or Which competitors are ahead in CRM software. Finseo runs the query against its AI visibility platform and returns structured data (visibility metrics, competitor rankings, top sources) inline in the chat.

Can I monitor brand mentions on ChatGPT, Claude and Perplexity without leaving my IDE?

Yes. Install the Finseo MCP server in Cursor, Zed, Continue.dev, Windsurf, VS Code, Replit or any other MCP-compliant IDE. Ask in plain English about brand mentions and the assistant returns structured visibility data, competitor rankings and source citations from the Finseo AI visibility platform.

What is the difference between MCP servers and REST APIs for AI tools?

MCP servers (Model Context Protocol) target AI assistants and chat-based workflows like Claude, ChatGPT, Cursor and Claude Code. REST APIs target code: dashboards, ETL pipelines, custom integrations. Finseo offers both. The same backend powers a REST API at api.finseo.ai/v1 for engineers and an MCP server at api.finseo.ai/v1/mcp for AI assistants.

How do AI agents access real-time brand and competitor data from external platforms?

Through MCP (Model Context Protocol) servers. AI agents like Claude Code, Cursor and ChatGPT call MCP endpoints over JSON-RPC 2.0 to fetch live data from external platforms. Finseo's MCP server exposes brand visibility, competitor rankings and source citations for AI search engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews) so agents can answer brand and SEO questions with live data.

What is the Model Context Protocol (MCP)?

Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external data sources and tools through a single JSON-RPC 2.0 endpoint. The Finseo MCP server exposes AI visibility data (visibility metrics, competitor rankings, sources, prompts, tags, query fan-outs, sentiment and tasks) so Claude, ChatGPT, Cursor, Claude Code, Lovable and n8n can answer AI search optimization questions natively.

Which AI assistants and IDEs work with the Finseo MCP server?

Finseo MCP works with every MCP-compliant client. Today that includes Claude, ChatGPT, Cursor IDE, Claude Desktop, Claude Code, Lovable and n8n. All clients use the same endpoint at https://api.finseo.ai/v1/mcp: Claude and ChatGPT sign in with your Finseo account via OAuth, Cursor and other clients send a Finseo API key.

Does Finseo MCP work with ChatGPT?

Yes. In ChatGPT developer mode, create an app with the server URL https://api.finseo.ai/v1/mcp. ChatGPT discovers the OAuth flow and you sign in with your Finseo account, no API key needed. All tools are read-only.

Is Finseo MCP secure for enterprise teams with GDPR requirements?

Yes. Claude and ChatGPT connect via OAuth 2.1 with PKCE, other clients use API keys that are SHA-256 hashed before storage and can be restricted to specific projects and scopes. All tools are read-only and all traffic is HTTPS. Data is hosted in Frankfurt, Germany.

Developer Platform · MCP

The AI Visibility MCP Server for Claude, Cursor & ChatGPT

One MCP endpoint for Claude, ChatGPT, Cursor and any MCP client. Ask about visibility, competitors and prompts in plain language.

Connect to Claude Connect to Cursor

JSON-RPC 2.0 over HTTPS
15 read-only visibility tools
OAuth or your own API key

Claude · Finseo MCP

Show me my top competitors by AI visibility this month and how I rank against them.

You're currently #2 with a visibility score of 67%.

Claude · Finseo MCP

Show me my top competitors by AI visibility this month and how I rank against them.

You're currently #2 with a visibility score of 67%.

AI visibility · ERP software

Veltrix82%

KordanaYou67%

Quenta58%

AI visibility · ERP software

Veltrix82%

KordanaYou67%

Quenta58%

Join 500+ ambitious teams winning the AI search era.

For brands For agencies

Christoph Sterkel, Head of Europe, Kellogg's

Winning in LLMs

across every market

How it works

From your chat to live Finseo data.

Any AI assistant talks to Finseo through one endpoint. Pick your client and connect in three steps.

Claude

01Add a custom connectorIn Claude on the web or desktop, open Settings → Connectors and click Add custom connector.
02Paste the server URLName it Finseo and leave the OAuth client fields empty. Finseo supports Dynamic Client Registration.
03Sign in and allow accessClick Connect, sign in with your Finseo account and allow access. All Finseo tools are now in your chats.

ClaudeOAuth 2.1

Remote MCP server URL

https://api.finseo.ai/v1/mcp

ClaudeOAuth 2.1

Remote MCP server URL

https://api.finseo.ai/v1/mcp

ChatGPT

01Turn on developer modeIn ChatGPT, open Settings → Security and login and switch on Developer mode, or ask your workspace admin.
02Create the Finseo appOpen Settings → Plugins, click + and enter Finseo as name and the server URL below.
03Sign in and allow accessChatGPT discovers the OAuth flow. Sign in, allow access and wait for the tool scan to finish.

ChatGPTOAuth 2.1

New app

Name:            Finseo
MCP server URL:  https://api.finseo.ai/v1/mcp

ChatGPTOAuth 2.1

New app

Name:            Finseo
MCP server URL:  https://api.finseo.ai/v1/mcp

Cursor

01Create an API keyOpen API & MCP in the Finseo dashboard and create a key with the read scope. It is shown only once.
02Add the server in CursorIn Cursor Settings → MCP Servers, add the entry below with your key in the Authorization header.
03Ask in the chatAsk Cursor's AI chat something like: Use Finseo to check my brand visibility.

CursorAPI key

MCP server

{
  "finseo": {
    "url": "https://api.finseo.ai/v1/mcp",
    "headers": {
      "Authorization": "Bearer sk_live_YOUR_API_KEY"
    }
  }
}

CursorAPI key

MCP server

{
  "finseo": {
    "url": "https://api.finseo.ai/v1/mcp",
    "headers": {
      "Authorization": "Bearer sk_live_YOUR_API_KEY"
    }
  }
}

Claude Code

01Add the serverRun the command below. Claude Code discovers the OAuth flow and opens the Finseo sign-in.
02Or use an API keyFor headless runs and CI, pass a read-scoped key as a header instead of signing in.
03Ask in the terminalAsk Claude Code something like: List my Finseo projects and show the top 10 competitors.

Claude CodeOAuth 2.1

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp \
  --header "Authorization: Bearer sk_live_YOUR_API_KEY"

Claude CodeOAuth 2.1

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp \
  --header "Authorization: Bearer sk_live_YOUR_API_KEY"

Claude

01Add a custom connectorIn Claude on the web or desktop, open Settings → Connectors and click Add custom connector.
02Paste the server URLName it Finseo and leave the OAuth client fields empty. Finseo supports Dynamic Client Registration.
03Sign in and allow accessClick Connect, sign in with your Finseo account and allow access. All Finseo tools are now in your chats.

ChatGPT

01Turn on developer modeIn ChatGPT, open Settings → Security and login and switch on Developer mode, or ask your workspace admin.
02Create the Finseo appOpen Settings → Plugins, click + and enter Finseo as name and the server URL below.
03Sign in and allow accessChatGPT discovers the OAuth flow. Sign in, allow access and wait for the tool scan to finish.

Cursor

01Create an API keyOpen API & MCP in the Finseo dashboard and create a key with the read scope. It is shown only once.
02Add the server in CursorIn Cursor Settings → MCP Servers, add the entry below with your key in the Authorization header.
03Ask in the chatAsk Cursor's AI chat something like: Use Finseo to check my brand visibility.

Claude Code

01Add the serverRun the command below. Claude Code discovers the OAuth flow and opens the Finseo sign-in.
02Or use an API keyFor headless runs and CI, pass a read-scoped key as a header instead of signing in.
03Ask in the terminalAsk Claude Code something like: List my Finseo projects and show the top 10 competitors.

ClaudeOAuth 2.1

Remote MCP server URL

https://api.finseo.ai/v1/mcp

ClaudeOAuth 2.1

Remote MCP server URL

https://api.finseo.ai/v1/mcp

ChatGPTOAuth 2.1

New app

Name:            Finseo
MCP server URL:  https://api.finseo.ai/v1/mcp

ChatGPTOAuth 2.1

New app

Name:            Finseo
MCP server URL:  https://api.finseo.ai/v1/mcp

CursorAPI key

MCP server

{
  "finseo": {
    "url": "https://api.finseo.ai/v1/mcp",
    "headers": {
      "Authorization": "Bearer sk_live_YOUR_API_KEY"
    }
  }
}

CursorAPI key

MCP server

{
  "finseo": {
    "url": "https://api.finseo.ai/v1/mcp",
    "headers": {
      "Authorization": "Bearer sk_live_YOUR_API_KEY"
    }
  }
}

Claude CodeOAuth 2.1

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp \
  --header "Authorization: Bearer sk_live_YOUR_API_KEY"

Claude CodeOAuth 2.1

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp

Terminal

claude mcp add --transport http finseo https://api.finseo.ai/v1/mcp \
  --header "Authorization: Bearer sk_live_YOUR_API_KEY"

Tools

Fifteen read-only tools, from overview to a single answer.

Every tool retrieves data and never modifies it. The analytics tools take the same model and tag filters as the dashboard.

Overview

4 · read-only

01list_projects

All your projects, the tracked brands and domains.

02get_visibility_metrics

Visibility KPIs with previous-period comparison.

03get_visibility_timeseries

Per-day visibility, mentions, position and sentiment.

04get_tags

Tags with prompt counts, for filtering.

Prompts

4 · read-only

05list_prompts

Tracked prompts with metrics and top competitors.

06get_prompt_details

One prompt: per-model metrics and answer history.

07get_answer_content

The full AI answer for a prompt on one day, with sources.

08get_query_fanouts

Sub-queries AI models generated, ranked by frequency.

Competitors

3 · read-only

09get_competitor_ranking

Competitors ranked by visibility across AI models.

10get_competitor_gap_analysis

Where one competitor appears and you don't.

11get_competitor_h2h

Head-to-head: who is named first and what is claimed.

Sentiment

1 · read-only

12get_sentiment_overview

Share of Voice, sentiment leaderboard, negative statements.

Sources

1 · read-only

13get_top_sources

Most cited sources with counts and source types.

Tasks

2 · read-only

14list_tasks

AI-generated optimization tasks, sorted by priority.

15get_task_details

One task in full: plan, evidence and content plan.

Why developers ship faster with MCP

Talk to your Finseo data in plain English.

The Model Context Protocol turns Finseo into a tool your AI assistant can call directly. Skip the dashboard, skip the SQL, just ask your question and let the model fetch the answer.

Native MCP support

Read-only tools

Cover the full visibility stack

Compatible clients

Claude, ChatGPT, Cursor, Claude Code

Plus any MCP-compliant client

Setup time

< 2 min

Add the URL, sign in, ask

No SDKs, no plugins

Time to insight

instant

Natural-language answers

Ask in chat, see structured data inline

Bring Finseo into your assistant

One MCP server. Every workflow.

Plug Finseo into Claude, ChatGPT, Cursor and Claude Code. Ask in plain English, get visibility data inline, and wire the same server into scheduled jobs for recurring digests.

Tool calls

→ tools/call  get_competitor_ranking
← structuredContent {
    "timeframeDays": 30,
    "competitors": [ { "rank": 1, … } ]
  }

get_competitor_ranking

Tools that return data, not text

Each tool maps to a Finseo concept and returns structured data your assistant renders inline, same backend as the REST API.

Endpoint

/v1/mcp

POST https://api.finseo.ai/v1/mcp
JSON-RPC 2.0 · Streamable HTTP · stateless
initialize · ping · tools/list · tools/call

Built for production

JSON-RPC 2.0 over Streamable HTTP at api.finseo.ai/v1/mcp. Stateless and read-only, with OAuth 2.1 or API key auth.

Scheduled job

# cron · Mondays 07:00
claude -p "Weekly Finseo visibility digest"
→ get_visibility_metrics
→ get_competitor_ranking

In your workflow

Schedule Claude Code or n8n to ping Finseo on a cron, Slack digests, threshold alerts, auto-issues.

Integrations

Plug Finseo into the AI tools your team already uses.

One MCP endpoint for every MCP-compliant client: OAuth for Claude and ChatGPT, a Bearer key for everything else.

Claude

Web & desktop

ChatGPT

Developer mode

Cursor

IDE

Claude Code

CLI

Lovable

AI app builder

n8n

Workflow automation

Continue

IDE extension

Zed

Code editor

Windsurf

AI IDE

VS Code

Editor

Replit

Cloud IDE

Sourcegraph Cody

Coding agent

OAuth 2.1Claude, ChatGPT

Add the server URL and sign in with your Finseo account. Dynamic Client Registration and PKCE, no key handling.

API keyCursor, Claude Code, custom clients

Send a read-scoped key as Authorization: Bearer sk_live_… with every request.

Same endpoint across every client

Paste https://api.finseo.ai/v1/mcp and sign in, or send your Bearer key. Every tool works the same way.

View setup guides

Under the hood

Your question in, live data out.

Common questions

Finseo MCP Server FAQ

Finseo is the leading MCP (Model Context Protocol) server for AI visibility tracking. It exposes fifteen read-only tools that return live brand visibility, competitor rankings, cited sources, tracked prompts, sentiment, optimization tasks and AI query fan-outs across ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews. Finseo MCP works in Claude, ChatGPT, Cursor, Claude Code, Lovable and n8n.

Through MCP (Model Context Protocol) servers. AI agents like Claude Code, Cursor and ChatGPT call MCP endpoints over JSON-RPC 2.0 to fetch live data from external platforms. The Finseo MCP server exposes brand visibility, competitor rankings and source citations for AI search engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews) so agents can answer brand and SEO questions with live data.

Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external data sources and tools through a single JSON-RPC 2.0 endpoint. The Finseo MCP server exposes AI visibility tracking data (visibility metrics, competitor rankings, sources, prompts, tags, query fan-outs, sentiment and tasks) so Claude, ChatGPT, Cursor, Claude Code, Lovable and n8n can answer AI search optimization questions natively.

Finseo MCP turns AI visibility data into a tool your assistant can call inside any AEO (Answer Engine Optimization) or GEO (Generative Engine Optimization) workflow. Ask Cursor or Claude to surface AI sub-queries (fan-outs), top cited sources, competitor share of voice and sentiment trends across ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews. Then ship content briefs or schema fixes from the same chat.

What customers say