Vectara

🚀 Vectara MCP Server

The Vectara MCP Server provides fast and reliable RAG for applications with proxy capabilities, reducing hallucinations through the MCP protocol and powered by Vectara's trusted RAG platform.

🔌 Compatible with Claude Desktop and any other MCP clients!

Vectara MCP is also compatible with any MCP clients.

The Model Context Protocol (MCP) is an open standard that enables AI systems to seamlessly interact with their data sources and tools, facilitating secure two-way connections.

Vectara-MCP offers fast and reliable Retrieval Augmented Generation (RAG) for any application with proxy capabilities. It reduces hallucinations through the MCP protocol and is powered by Vectara's trusted RAG platform.

🚀 Quick Start

📦 Installation

You can install the package directly from PyPI:

pip install vectara-mcp

✨ Features

Available Tools

• ask_vectara: Run a RAG query using Vectara and return search results with a generated response.

  Parameters:

  • query: str, User query - Required.
  • corpus_keys: list[str], List of Vectara corpus keys to use - Required. Ask the user to provide one or more corpus keys.
  • api_key: str, Vectara API key - Required.
  • n_sentences_before: int, Number of sentences before the answer to include in the context - Optional, default is 2.
  • n_sentences_after: int, Number of sentences after the answer to include in the context - Optional, default is 2.
  • lexical_interpolation: float, Amount of lexical interpolation to use - Optional, default is 0.005.
  • max_used_search_results: int, Maximum number of search results to use - Optional, default is 10.
  • generation_preset_name: str, Name of the generation preset to use - Optional, default is "vectara-summary-table-md-query-ext-jan-2025-gpt-4o".
  • response_language: str, Language of the response - Optional, default is "eng".

  Returns:

  • Vectara's response, including the generated answer and search results.

• search_vectara: Run a semantic search query using Vectara without generation.

  Parameters:

  • query: str, User query - Required.
  • corpus_keys: list[str], List of Vectara corpus keys to use - Required. Ask the user to provide one or more corpus keys.
  • api_key: str, Vectara API key - Required.
  • n_sentences_before: int, Number of sentences before the answer to include in the context - Optional, default is 2.
  • n_sentences_after: int, Number of sentences after the answer to include in the context - Optional, default is 2.
  • lexical_interpolation: float, Amount of lexical interpolation to use - Optional, default is 0.005.

  Returns:

  • Vectara's semantic search results.

💻 Usage Examples

Configuration on Claude Desktop

Here is an example of how to configure the Vectara MCP Server on Claude Desktop:

{
  "VectaraMCP": {
    "enabled": true,
    "apiKey": "your_api_key_here",
    "corpusKeys": ["corpus1", "corpus2"],
    "nSentencesBefore": 2,
    "nSentencesAfter": 2,
    "lexicalInterpolation": 0.005
  }
}

Usage in Claude Desktop App

After installation, you need to restart Claude Desktop to apply the configuration. Then you can call the tools like this:

# Query using the ask_vectara tool
python -m vectara_mcp.ask_vectara --query "Your query"

# Semantic search using search_vectara
python -m vectara_mcp.search_vectara --query "Your query"

Vectara RAG Examples

Here are examples of how to use Vectara's RAG capabilities for queries:

Basic Usage - Using the ask_vectara Tool

$ python -m vectara_mcp.ask_vectara --query "How to install a Python library?"

Advanced Usage - Using the search_vectara Tool

$ python -m vectara_mcp.search_vectara --query "Python installation guide"

These examples demonstrate how to leverage Vectara's powerful features for information retrieval and generation.