--- title: "Axiom MCP: review, setup & limitations · Undominated.ai" canonical: https://undominated.ai/mcp-servers/axiom/ description: "Queries Axiom telemetry and manages observability resources through an account-authorized hosted connection." --- # Axiom MCP: review, setup & limitations · Undominated.ai > Queries Axiom telemetry and manages observability resources through an account-authorized hosted connection. [← Explore all mcp servers](/mcp-servers/) OBSERVABILITY / Axiom # Axiom MCP Queries Axiom telemetry and manages observability resources through an account-authorized hosted connection. [See setup guidance ↓](#setup)[Original source ↗](https://axiom.co/docs/console/intelligence/mcp-server) SOURCE REVIEW Reviewed 2026-10-14 Evidence 2 linked sources Publisher Axiom Licence Not established [Read what was—and wasn’t—checked ↓](#review) “Connect AI agents to Axiom with the Model Context Protocol (MCP) Server.” [Axiom · upstream description ↗](https://axiom.co/docs/console/intelligence/mcp-server) Our analysis follows below. 01 / THE REASONING ## Why this made the selection. - Exposes APL queries and dataset exploration alongside observability resource tools. - Uses the connected user’s role and permissions rather than granting access to all organization data. - Axiom’s page describes a remote MCP server that uses OAuth so agents do not see the user’s tokens, with an API-token header path for agents that cannot use browser OAuth. - Agents that do not support remote MCP are told to use a separate bridge, and the tools page only lists the exposed tools and prompts. ### A good fit for - Investigating logs and metrics in an authorized Axiom organization - Querying an Axiom organization from a remote MCP client that can complete OAuth - A header-based API token only when the agent cannot use browser OAuth ### Weigh up before choosing - Tools also manage dashboards, monitors, notifiers and datasets; this is not a read-only connector. - The documented MCP service runs in US infrastructure even when a dataset uses another region. Review that processing path for regional workloads. - Queries and tool outputs can expose sensitive telemetry and consume the connected account’s query allowance. - The page says agents never see the OAuth tokens. It does not say a pasted API token is hidden the same way. - An agent that does not support remote MCP is not told to call the remote URL directly. - No licence was stated on these pages, and none was inferred. - This review did not execute the resource. - The server was not called. - The hosted routing region matters for telemetry handling; self-hosted operation is a separate deployment, and the local server does not supply MCP prompts. 02 / THE REVIEW RECORD ## What we actually inspected. Source review has boundaries. A clear record is more useful than a “safe” badge. ### Material inspected - Official MCP setup, data-flow and permissions documentation - Official tool reference - https://axiom.co/docs/console/intelligence/mcp-server - https://axiom.co/docs/console/intelligence/mcp-server/tools - Round 2 reread of acquired publisher setup and permission evidence; see immutable wave review. ### Our findings - The current tools include both query and management actions; regional processing and account permissions are disclosed. - The remote Axiom MCP server uses OAuth, and the page says that flow keeps credentials isolated so agents never see the tokens. - If the agent does not support browser OAuth, the page says to authenticate with headers and an API token instead of OAuth 2.0. - If the agent does not support remote MCP directly, the page points elsewhere rather than at the remote server alone. The tools page describes the exposed tools and prompts, including a way to send feedback about missing capabilities. - The publisher says hosted MCP query results route through US infrastructure, including queries against a remote data deployment. ### Not established by this review - No MCP server or upstream package was installed or run. - No authenticated account, client handshake, query, write operation or security test was performed. - The resource was not executed and its tools were not called. - The server was not called. A source review is not a runtime or security certification. The review applies to the material and revision named here. A newer upstream release can change its behavior. 03 / PUT IT TO WORK ## Connect a server deliberately. [Upstream setup instructions ↗](https://axiom.co/docs/console/intelligence/mcp-server) - Add https://mcp.axiom.co/mcp as a remote HTTP MCP server. - Complete OAuth with the intended Axiom organization and review the user’s role and dataset permissions. - Start with a narrow query; approve any resource-management action separately. ### Before you start - An Axiom account with the relevant dataset and resource permissions - An OAuth-capable remote MCP client ### Documented configuration example Use the client and authentication context described in the setup guidance. Configuration formats vary between hosts. { "servers": { "axiom": { "type": "http", "url": "https://mcp.axiom.co/mcp" } } } Copy configuration ↗ ### Compatibility Remote MCP clients ### Transports Streamable HTTP ### Hosted account connection - Read allowed datasets and run APL queries - Create, update or delete supported observability resources where the user role permits ### Cost model Hosted Axiom service under account and query-usage terms; assistant usage is separate. No open-source licence is asserted for this hosted implementation. 04 / FOLLOW THE EVIDENCE ## The source trail. Our notes are separate from the original resource. Check upstream before adopting a new version. - [Official setup and capabilities ↗](https://axiom.co/docs/console/intelligence/mcp-server) Checked 2026-10-14 https://axiom.co/docs/console/intelligence/mcp-server Supports: summary, upstreamDescription, whySelected, bestFor, install, compatibility, transports, limitations, access, review - [Axiom tool reference ↗](https://axiom.co/docs/console/intelligence/mcp-server/tools) Checked 2026-10-14 https://axiom.co/docs/console/intelligence/mcp-server/tools Supports: review, limitations, access KEEP COMPARING ## Other approaches to consider. Related by category or shared topics. These are alternatives to inspect, not a measured quality order. [### Datadog MCP ↗ Query Datadog telemetry and operate permitted monitoring resources from an authenticated AI client.](/mcp-servers/datadog/)[### Checkly MCP Server ↗ Checkly's hosted MCP server over Streamable HTTP: inspect check status and results, test sessions, RCA and result assets, trigger deployed checks, manage monitors and groups, environment variables and status-page incidents. OAuth is recommended; user and service API keys are the documented alternatives.](/mcp-servers/checkly/)[### Logfire MCP ↗ Connects assistants to Logfire traces, metrics and observability management tools in the chosen region.](/mcp-servers/logfire/) [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/) ## Continue your investigation - [Inspect resource evidence](/tools/) - [Choose a host](/agents/) - [Inspect reusable workflows](/skills/) - [Use Undominated data](/api/)