Axiom MCP Server

**This MCP server is not part of the Databricks Services, and its use is subject to the source's license terms. The MCP server and the information below are reproduced here for your convenience from publicly available listings. Any issues should be submitted to the MCP provider.**

**Overview**

The Axiom MCP Server connects AI assistants to your Axiom observability data, enabling agents to query logs, metrics, traces, and events using the Axiom Processing Language (APL). Purpose-built for wide schemas and high-cardinality data, the server outputs CSV-formatted tabular data instead of verbose JSON, applies automatic time-series bucketing (`maxBinAutoGroups`), and uses adaptive truncation to stay within LLM context limits. The result: agents can handle more queries per session without hitting token ceilings.

Available as a hosted remote server at `https://mcp.axiom.co/mcp` or as a self-hosted Go binary. The hosted server supports runtime tuning via URL parameters — `?max-age=500` to cap result size, `?with-otel=1` to enable the OpenTelemetry tool family. Only a minimal core toolset is exposed by default to keep the initial handshake light.

The repository includes a Cloudflare Workers app (the hosted server) and a TypeScript package with core MCP utilities.

**Tools**

- `queryApl` — Execute APL queries against datasets with automatic result shaping
- `listDatasets` — List available datasets
- `getDatasetSchema` — Retrieve schema for a specific dataset
- `getSavedQueries` — Retrieve saved/starred APL queries
- `getMonitors` — List monitoring configurations
- `getMonitorsHistory` — Get monitor execution history

**Use Cases**

- **Incident investigation:** Query recent error logs, correlate across services, and summarize root cause findings.
- **Ad-hoc observability:** Generate queries and summaries from log and trace data without writing APL manually.
- **Monitor review:** Inspect monitoring configurations and alert history through conversational prompts.
- **AI trace analysis:** Review agent workflows, evaluate outputs, and track cost/latency across LLM providers.

**Learn More:** [github.com/axiomhq/mcp](https://github.com/axiomhq/mcp)