Axiom

You can run a JavaScript-based MCP server that lets AI agents query data through Axiom Processing Language (APL). It’s packaged to integrate smoothly with Node.js environments, and you can start it quickly using a simple MCP configuration or a global install. This guide shows practical steps to install, configure, and run the server, plus how to use its endpoints to query data and list datasets.

How to use

You run the MCP server as a local process and connect your MCP client to it. The server exposes endpoints to inspect its implementation, enumerate available tools, and invoke specific tools. The main tools you’ll use are a query tool for APL queries and a dataset listing tool. Start by ensuring your token and organization details are correctly set up, then query APL data or retrieve the available datasets. When you call a tool, you provide the necessary arguments, and the server returns the results you can feed back into your AI agent workflows.

How to install

Prerequisites you need on your system are Node.js and npm (Node Package Manager). You will install the MCP server package globally, configure environment variables, and run the server using the CLI provided by the package.

Step by step commands you can run now:

# Install the MCP server package globally
npm install -g mcp-server-axiom

# Optional: install locally in a project instead of global install
# npm install mcp-server-axiom

# Start the server using a config file (example path):
mcp-server-axiom config.json

Additional content

Configuration is provided through an MCP configuration entry that defines how to run the server and which environment variables to supply. The server relies on Axiom tokens and organization IDs to authorize requests, along with optional URL and rate-limiting settings.

Example MCP configuration to run the Axiom MCP server via npx with token and org details is shown below.

{
  "axiom": {
    "command": "npx",
    "args": ["-y", "mcp-server-axiom"],
    "env": {
      "AXIOM_TOKEN": "<AXIOM_TOKEN_HERE>",
      "AXIOM_URL": "https://api.axiom.co",
      "AXIOM_ORG_ID": "<AXIOM_ORG_ID_HERE>"
    }
  }
}

Local development and testing can be done by installing the package globally, setting environment variables, and launching the server. The following environment variables are commonly used.

export AXIOM_TOKEN=your_token
export AXIOM_ORG_ID=your_org_id
export AXIOM_URL=https://api.axiom.co
export AXIOM_QUERY_RATE=1
export AXIOM_QUERY_BURST=1
export AXIOM_DATASETS_RATE=1
export AXIOM_DATASETS_BURST=1
export PORT=3000

mcp-server-axiom

If you prefer a configuration file, you can provide the settings in JSON as shown in the example.

{
  "token": "your_token",
  "url": "https://custom.axiom.co",
  "orgId": "your_org_id",
  "queryRate": 2,
  "queryBurst": 5,
  "datasetsRate": 1,
  "datasetsBurst": 2
}

Environment and endpoints

The server exposes a small HTTP API with endpoints to retrieve server info and to call tools. You can call the APL query tool or list datasets through the provided tool endpoints.

Available tools

queryApl

Execute APL queries against Axiom datasets and return results to your MCP client.

listDatasets

List available datasets from your Axiom account.