Can AI agents automate Axiom?

Axiom is automatable today through the official hosted MCP server. This record is the Axiom observability platform for event data queried with APL, including datasets, metrics, dashboards, monitors, and notifiers. Datadog, Grafana, Honeycomb, and SigNoz are separate products. The recommended MCP endpoint is https://mcp.axiom.co/mcp with browser OAuth. Clients that cannot use OAuth send Authorization: Bearer with a personal access token and x-axiom-org-id. Legacy SSE is https://mcp.axiom.co/sse. Documented tools query datasets and metrics, list and create datasets, and create, update, and delete dashboards, monitors, and notifiers. Query results from the remote server are routed through US infrastructure. The REST API uses https://api.axiom.co for non-ingest calls, JSON bodies, and either an API token or a personal access token plus the organization id. API tokens can be limited to actions such as querying one dataset. The Axiom CLI installs with go install github.com/axiomhq/cli/cmd/axiom@latest and is documented for ingest, authentication state, and multiple organizations. Connector catalogs were not opened. https://axiom.co/llms.txt is a real documentation index. Computer-use viability is unassessed. MCP scores 8 and is the best path because agents can query and manage monitors without assembling ingest and query calls themselves. The REST API scores 7. The CLI scores 5 because the opened page covers ingest and auth rather than the full console.

Best path today: mcp · Overall automatability: 8/10