Hermes Search

Hermes Search MCP enables full-text and semantic search over Azure Cognitive Search with type-safe operations, powered by the open Model Context Protocol (MCP).

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "cognitive-stack-hermes-search-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "hermes-search-mcp@latest"
      ],
      "env": {
        "AZURE_SEARCH_API_KEY": "your-api-key",
        "AZURE_SEARCH_ENDPOINT": "your-search-endpoint",
        "AZURE_SEARCH_INDEX_NAME": "your-index-name"
      }
    }
  }
}

Hermes Search MCP Server provides seamless, secure access to full-text and semantic search over structured and unstructured data via the Model Context Protocol. It connects with Azure Cognitive Search, enabling you to index content, run rich search queries, and manage your search index from MCP clients with type-safe TypeScript support.

How to use

You use the Hermes Search MCP Server by connecting your MCP client to the local or remote server and performing search, index, and management actions against your Azure Cognitive Search index. After you install and configure the server in your client, you can search for documents containing specific phrases, push new documents into the Azure index, and delete or manage the index as needed. The workflow is designed to be practical and secure, with credentials kept in environment variables and exposed only to the server during operation.

How to install

Prerequisites you need before installing the Hermes Search MCP Server:

Node.js (v20 or higher)
Azure Cognitive Search service and credentials
Git (only needed if you plan to install via Git)

Install the server using NPX for a quick setup:

npx -y hermes-search-mcp@latest

If you want to install Hermes Search MCP Server for a specific MCP client via Smithery, run this command:

npx -y @smithery/cli install @hermes-search/mcp --client claude

Additional sections

Configuration and environment variables are used to securely connect to Azure Cognitive Search. Ensure you provide valid values for the search endpoint, API key, and index name in your client configuration. After setting up, restart your MCP client to load the Hermes Search MCP Server.

Troubleshooting covers common issues such as verifying installation, checking credentials, and ensuring the target Azure index exists and is accessible. If you encounter problems, confirm your Node.js version, validate your environment variables, and review the client’s MCP configuration syntax.

Usage examples demonstrated by the Hermes Search MCP Server include searching for documents by keywords, indexing content into Azure Cognitive Search, and deleting the current index when needed.

Available tools

searchDocuments

Query the Azure Cognitive Search index for documents matching a query, returning relevant results with configurable limits and facets.

indexContent

Index new documents into Azure Cognitive Search by providing document objects with fields such as id, title, and content.

deleteIndex

Delete the current Azure Cognitive Search index to reset or reprepare the data store for new indexing.

Built by

VeilStrat

AI signals for GTM teams