Agentic Astra

Astra MCP Server

python

1

GitHub Stars

python

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": {
    "smatiolids-agentic-astra": {
      "command": "uvx",
      "args": [
        "agentic-astra",
        "--env-file",
        ".env",
        "--tr",
        "http|sse|stdio"
      ],
      "env": {
        "ASTRA_TOKEN": "YOUR_TOKEN"
      }
    }
  }
}

Agentic Astra provides a semantic layer to connect AI Agents to Astra DB through MCP, enabling you to translate structured and unstructured data with configurable business logic and to interact through a dedicated UI for tool editing and tool generation.

How to use

You connect your MCP client to the Agentic Astra MCP server to access tools that interact with Astra DB. Tools are loaded from a collection in Astra DB or from a file, transformed into function definitions, and exposed for use inside an agent workflow. When a tool is invoked, the server translates the call into Astra DB or DataStax HCD operations, applies the necessary filters, and returns results to your MCP client. If embedding or similarity search is needed, you can leverage models from OpenAI or IBM Watsonx.

How to install

Prerequisites You need Node.js installed (LTS version recommended) and access to Astra DB with an application token. You will also work with a tool definitions collection stored in Astra DB or a local tool definitions file.

Step 1: Set up environment variables Create a .env file based on the provided example and fill in the required values such as Astra DB credentials and tokens.

Step 2: Install or prepare the MCP server dependencies Depending on your workflow, you will use the MCP runtime to start the server locally and optionally access the UI for tool creation.

Step 3: Start the MCP server Locally with HTTP You can start the server for HTTP access using the following command.

uv run agentic-astra --env-file .env -tr http

Copy the .env.example file to .env and fill the variables.

Additional setup and tool creation

Create tools that the MCP server will expose by using the Agentic Astra UI. Run the UI locally to define tools and save them to your Astra DB collection.

npx agentic-astra-ui --env-file .env

Run variants and debugging

Run the server with a development log level to troubleshoot and verify behavior. You can start in HTTP mode or switch to a streaming interface if your workflow requires it.

uv run agentic-astra --env-file .env --log-level debug -tr http

Using MCP Inspector (STDIO)

For interactive inspection of MCP activity, you can run the inspector against the STDIO version of the server to observe calls and responses as you exercise tools.

npx @modelcontextprotocol/inspector uv run agentic-astra --log-level debug -tr stdio --env-file .env