# Vee3 [Health: Active]

**Category:** 🌐 Social Media  
**Repository:** https://github.com/Vee3io/Vee3  
**GitHub Stars:** 1  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/vee3

## Description
Hosted MCP with 78 agent tools: X, domains, SEO, Trends, YouTube, TikTok, screenshots, and more.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "vee3": {
    "url": "https://mcp.vee3.io/mcp"
  }
}
```

## Documentation

## What the Vee3 MCP server does

The Vee3 MCP server gives AI agents access to a hosted collection of live-data and action tools through a single MCP connection. The repository description lists 78 tools spanning research, social platforms, search data, media metadata, website analysis, file handling, and Clerk application data.

The available areas include domain availability research; X post searches; Google Trends and Google Search Insights; SEO and backlink data; Google Maps place searches; YouTube and TikTok data; Rotten Tomatoes metadata; website screenshots; file uploads and downloads; and discovery helpers for finding tools in the catalog. Clerk-related tools can work with users, organizations, sessions, and other data in the user's own Clerk application.

Typical tasks include checking whether a domain is available, finding recent posts about a topic on X, comparing search trends, reviewing a website screenshot, researching videos by views or engagement, comparing Rotten Tomatoes scores, or listing users from a Clerk application.

## How it works

Vee3 exposes the catalog through the hosted MCP endpoint `https://mcp.vee3.io/mcp`. An MCP-compatible client connects to that URL and sends the Vee3 API key as a request header. The service then makes the selected catalog tools available to the agent.

The Vee3 MCP server is intended for tasks that depend on current or externally sourced information, including market research, due diligence, competitive analysis, and other fact-heavy workflows. It can be used from an editor or another MCP client without requiring each data source to be configured as a separate local server, according to the provided description.

## Setup and configuration

Create an account at `app.vee3.io`, generate an API key from the dashboard's keys area, and configure the hosted MCP connection with that key. The required environment variable is `VEE3_API_KEY`, although the README specifically describes passing the credential as a request header on the hosted connection.

The service has a free tier. Tool calls use token-based billing, and failed calls are not charged. The provided material does not specify the token rates, request-header name, or client-specific configuration format, so those details should be taken from Vee3's documentation before deployment.

## Tools and capabilities

The catalog supports workflows such as:

- Searching X for recent posts about specified subjects.
- Finding domains and investigating SEO or backlink information.
- Exploring Google Trends and search insights.
- Looking up places through Google Maps.
- Searching YouTube and TikTok content and examining listed engagement data.
- Retrieving Rotten Tomatoes metadata and comparing critic and audience scores.
- Capturing website screenshots for design or usability review.
- Uploading and downloading files.
- Reading and managing supported data in a user's Clerk application.
- Discovering available tools in the Vee3 catalog.

The exact inputs, outputs, and limits for individual tools are documented on Vee3's capabilities page.

## Limitations and notes

Vee3 requires an API key and is not a credential-free local server. Access to some capabilities may also depend on the connected service or the user's own account, such as Clerk application data. Pricing is usage-based after or beyond the free tier, but the supplied material does not define the included allowance or token prices.

The README does not identify specific desktop or coding clients by name. It also does not provide a local package, Docker image, or local installation command; the documented connection method is the hosted MCP endpoint.

_Full upstream README: https://allmcps.com/mcp/vee3/readme_