Valyu MCP Server

Name: Valyu MCP Server
Author: valyu-network

@valyu-network

リポジトリ

Valyu MCP Server について

Typescript implementation of MCP server for Valyu Network API (https://docs.valyu.network/api-reference)

#valyu #mcp #server #api #knowledge-retrieval

基本情報

カテゴリ

その他

ライセンス

MIT

ランタイム

node

トランスポート

stdio

公開者

valyu-network

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "valyu-mcp-js": {
      "command": "docker",
      "args": [
        "pull",
        "ghcr.io/tiovikram/valyu-mcp-server"
      ]
    }
  }
}

ツール

Search proprietary and/or web sources for information

Submit user feedback for a transaction

概要

What is Valyu MCP Server?

Valyu MCP Server is a Model Context Protocol server that provides access to Valyu's knowledge retrieval and feedback APIs. It enables LLMs to search proprietary and web sources for information and submit feedback on transactions.

How to use Valyu MCP Server?

Install via Docker (docker pull ghcr.io/tiovikram/valyu-mcp-server), set the VALYU_API_KEY environment variable, and configure the server in your MCP client (e.g., Claude.app). Invoke the knowledge tool to search sources or the feedback tool to submit user feedback.

Key features of Valyu MCP Server

knowledge tool to search proprietary, web, or all sources
feedback tool to submit user feedback for transactions
Supports configurable search parameters (query, max price, data sources, etc.)
Sentiment ratings: very good, good, bad, very bad
Docker image ready for easy deployment

Use cases of Valyu MCP Server

Searching for information across proprietary and web data sources
Submitting user feedback and sentiment for a specific transaction
Integrating knowledge retrieval directly into an LLM workflow

FAQ from Valyu MCP Server

What tools are available?

Two tools: knowledge for searching information and feedback for submitting user feedback on transactions.

What arguments are required for the knowledge tool?

A query (string), search_type ("proprietary", "web", or "all"), and max_price (number, maximum CPM allowed). Optional arguments include data_sources, max_num_results, similarity_threshold, and query_rewrite.