Vector GraphQL MCP Server

What is Vector GraphQL MCP Server?

Vector GraphQL MCP Server is an MCP server that provides tools for querying the Vector GraphQL API. It is intended for developers who want to retrieve leaderboard data, trader profiles, trending Solana tokens, and token broadcasts from Vector.

How to use Vector GraphQL MCP Server?

Clone the repository, install dependencies with uv or pip, then run in development mode using uv run mcp dev vector_server.py or install in Claude Desktop with uv run mcp install vector_server.py.

Key features of Vector GraphQL MCP Server

• Query Vector leaderboard data via fetch_leaderboard
• Retrieve detailed trader profiles via fetch_profile
• Get trending Solana tokens via fetch_token_data
• Fetch broadcasts for specific tokens via fetch_token_broadcasts
• Supports installation with uv or pip
• Includes MCP Inspector for development testing

Use cases of Vector GraphQL MCP Server

• Monitor top traders on Vector by fetching leaderboard rankings
• Analyze individual trader performance by fetching profile details
• Track trending Solana tokens with recent broadcast activity
• Retrieve token-specific broadcast history for research or alerts

FAQ from Vector GraphQL MCP Server

How do I set up the server?

Clone the repository, install dependencies with uv add "mcp[cli]" httpx or pip install -r requirements.txt, then run uv run mcp dev vector_server.py or mcp dev vector_server.py.

What dependencies are required?

The server requires mcp[cli] and httpx when using uv, or the packages listed in requirements.txt when using pip. For CLI tools separately, install mcp, httpx, click>=8.0, rich>=10.0, and typer>=0.9.0.

What tools are available?

Four tools: fetch_leaderboard, fetch_profile, fetch_token_data, and fetch_token_broadcasts. Each has documented parameters (e.g., leaderboard_type for fetch_leaderboard, username for fetch_profile, token_id for fetch_token_broadcasts; fetch_token_data has no parameters).

How can I install the server in Claude Desktop?

Run uv run mcp install vector_server.py (with uv) or mcp install vector_server.py (with pip).