Skip to content

Summary

Axiom's official MCP server — query event data with APL, read dataset schemas, and create or edit dashboards, monitors and notifiers from an agent.

Features

  • queryApl runs APL against your datasets; getDatasetSchema keeps the queries valid
  • Metrics tools: listMetrics, searchMetrics, queryMetrics, tags and tag values
  • Create, update, export and delete dashboards, including single-chart updates
  • Create and update monitors and notifiers, and read monitor history
  • Remote at https://mcp.axiom.co/mcp with OAuth 2.0; SSE endpoint also available
  • One-click Cursor installer; Claude Code and Claude Desktop config documented

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it.
  3. Restart the agent — the "Axiom" tools will be available.
{
  "mcpServers": {
    "axiom-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.axiom.co/mcp"
      ]
    }
  }
}

Description

Axiom is an event-data platform: logs, traces and metrics land in datasets you query with APL, Axiom's own query language. The official Axiom MCP server hands an agent the same surface, so "what changed in the checkout service between 14:00 and 15:00?" becomes an APL query the agent writes, runs and reads back — with the dataset schema available so it can write a query that actually parses.

Remote and OAuth-backed. The server is at https://mcp.axiom.co/mcp (an SSE endpoint exists at https://mcp.axiom.co/sse). Remote connections authenticate with OAuth 2.0; a local run can use an API token instead. Claude Code takes the URL as a remote MCP server, Cursor has a one-click installer, and any client that can shell out uses npx -y mcp-remote https://mcp.axiom.co/mcp.

Twenty-four tools, and a good half of them write. Reads: queryApl, listDatasets, getDatasetSchema, getSavedQueries, listMetrics, searchMetrics, queryMetrics, listMetricTags, getMetricTagValues, listDashboards, getDashboard, exportDashboard, checkMonitors, getMonitorHistory, listNotifiers. Writes: creating, updating and deleting dashboards, monitors and notifiers, plus updateDashboardChart.

That write half is what separates this from a read-only telemetry bridge: an agent that has just found a regression can open the monitor for it and wire up the notifier in the same turn, rather than handing you a query to paste. Scope the credential accordingly — the tools are as powerful as the token behind them.

Related MCP Servers

Checkly's hosted MCP server gives an agent read access to your synthetic monitoring account — check health, failing results, test sessions, Rocky AI root-cause analyses, incidents and status pages — over streamable HTTP with OAuth.

MCP: Grafana

by Grafana Labs

Featured

Official Grafana MCP server: search and edit dashboards, query Prometheus, Loki and Pyroscope, and work with alerts, incidents and on-call from an agent.

Browse all MCP servers →