MCP.so
ログイン
V

Voyageai Cli

@mrlynn

リポジトリ

Voyageai Cli について

MCP server for Voyage AI embeddings, reranking, and MongoDB Atlas Vector Search. Provides 11 tools for semantic search, document ingestion, cost estimation, and model exploration. Full RAG pipeline: chunk, embed, vector search, and rerank from any MCP client.

#voyage-ai#embeddings#vector-search#mongodb#rag#reranking#semantic-search#retrieval

基本情報

カテゴリ

メモリとナレッジ

トランスポート

stdio

公開者

mrlynn

投稿者

Michael Lynn

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": {
    "vai": {
      "command": "npx",
      "args": [
        "voyageai-cli",
        "mcp-server"
      ],
      "env": {
        "VOYAGE_API_KEY": "<YOUR_VOYAGE_API_KEY>"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Voyageai Cli?

An MCP server that gives AI agents access to Voyage AI's embedding and reranking models plus MongoDB Atlas Vector Search. It is designed for building RAG pipelines directly from an AI agent without leaving the agent interface.

How to use Voyageai Cli?

Install globally with npm install -g voyageai-cli, then run vai mcp install claude-code or vai mcp install all to configure the server for your agent. Alternatively, use npx without installation. Requirements: Node.js 20+, a Voyage AI API key (free at dash.voyageai.com), and a MongoDB Atlas cluster for retrieval/ingestion tools (utility tools work without it).

Key features of Voyageai Cli

  • 11 tools covering retrieval, embedding, management, utility, and ingestion
  • Full RAG pipeline: vai_query, vai_search, vai_rerank
  • Embedding generation with vai_embed and cosine similarity with vai_similarity
  • List indexes (vai_collections) and browse models (vai_models)
  • Learning tools: vai_topics, vai_explain, and cost calculator vai_estimate
  • Ingestion pipeline: chunk, embed, and store documents via vai_ingest

Use cases of Voyageai Cli

  • Build end-to-end RAG applications from within an AI agent
  • Perform vector search on MongoDB Atlas collections
  • Rerank search results for improved relevance
  • Generate embeddings for custom data pipelines
  • Estimate costs before large-scale ingestion

FAQ from Voyageai Cli

What dependencies and runtime does Voyageai Cli require?

Node.js version 20 or later, a Voyage AI API key (free at dash.voyageai.com), and optionally a MongoDB Atlas cluster for retrieval and ingestion tools.

How do I install Voyageai Cli?

Run npm install -g voyageai-cli, then use vai mcp install claude-code or vai mcp install all to connect it to your agent. You can also use npx without installing.

What tools are available in Voyageai Cli?

Eleven tools: vai_query, vai_search, vai_rerank (retrieval); vai_embed, vai_similarity (embedding); vai_collections, vai_models (management); vai_topics, vai_explain, vai_estimate (utility); and vai_ingest (ingestion).

Where can I find documentation and support?

Documentation is at https://github.com/mrlynn/voyageai-cli/blob/main/docs/mcp-server.md. General information is at https://vai.mlynn.org. The npm package is at https://www.npmjs.com/package/voyageai-cli.

What transport does Voyageai Cli use?

The README does not specify transport details. It indicates integration via MCP (Model Context Protocol) commands for standard agent communication.

コメント

「メモリとナレッジ」の他のコンテンツ

Docs MCP Server

arabold

Grounded Docs MCP Server: Open-Source Alternative to Context7, Nia, and Ref.Tools

1.5K

Obsidian MCP Server

StevenStavrakis

A simple MCP server for Obsidian

717

Semantic Scholar MCP Server

YUZongmin

A FastMCP server implementation for the Semantic Scholar API, providing comprehensive access to academic paper data, author information, and citation networks.

158

Obsidian MCP Server

cyanheads

Read, write, search, and surgically edit Obsidian vault notes, tags, and frontmatter via MCP. STDIO or Streamable HTTP.

618

Anytype MCP Server

anyproto

An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.

467