MCP tools reference

For setup, see Axiom MCP Server.

Tools at a glance#

Query#

• queryDataset (Read, path) — Run an APL query against an events or traces dataset. For metrics datasets, use queryMetrics instead.

  Example: “Show the top 10 status codes in logs for the last 30 minutes.”

• getSavedQueries (Read, path) — List saved APL queries in your organization.

  Example: “List my saved queries.”

Datasets#

• listDatasets (Read, path) — List available Axiom datasets.

  Example: “List datasets.”

• getDatasetFields (Read, path) — List fields in an events or traces dataset, including field names and types. For metrics datasets, use listMetrics and listMetricTags instead.

  Example: “List the fields in the dataset logs.”

• createDataset (Write, path) — Create a new dataset in the current organization.

  Example: “Create an events dataset named app-logs.”

Metrics#

• listMetrics (Read, path) — List available metric names in a dataset.

  Example: “List metrics in the metrics dataset.”

• listMetricTags (Read, path) — List tags available for filtering metrics in a dataset.

  Example: “What tags can I filter on in the metrics dataset?”

• getMetricTagValues (Read, path) — Get values for a specific metric tag in a dataset.

  Example: “List values for the service.name tag in metrics.”

• searchMetrics (Read, path) — Search for metrics that match a known tag value.

  Example: “Find metrics tagged with service.name = api-gateway.”

• queryMetrics (Read, path) — Run an MPL query against an Axiom metrics dataset.

  Example: “Show CPU usage for api-gateway over the last hour.”

Dashboards#

• listDashboards (Read, path) — List dashboards in your organization.

  Example: “List my dashboards.”

• getDashboard (Read, path) — Get details for a dashboard by UID.

  Example: “Show the configuration for dashboard abc123.”

• exportDashboard (Read, path) — Export a dashboard as JSON.

  Example: “Export dashboard abc123 as JSON.”

• createDashboard (Write, path) — Create a dashboard from a JSON document.

  Example: “Create a dashboard that charts error rate for logs over the last 24 hours.”

• updateDashboard (Write, path) — Update a dashboard by UID.

  Example: “Update dashboard abc123 to add a panel for p99 latency.”

• updateDashboardChart (Write, path) — Patch a single chart in an existing dashboard by chart ID.

  Example: “Change the time range on chart chart-1 in dashboard abc123 to the last 6 hours.”

• deleteDashboard (Write, path) — Delete a dashboard by UID.

  Example: “Delete dashboard abc123.”

Monitors and notifiers#

• checkMonitors (Read, path) — Check monitors and their current statuses.

  Example: “Which monitors are currently alerting?”

• getMonitor (Read, path) — Get a monitor’s full definition by ID, including the complete query text.

  Example: “Show the full configuration for monitor mon_123.”

• getMonitorHistory (Read, path) — Get alert history for a specific monitor.

  Example: “Show alert history for monitor mon_123 over the last 7 days.”

• listNotifiers (Read, path) — List configured notifiers.

  Example: “List notifiers.”

• createMonitor (Write, path) — Create a new monitor.

  Example: “Create a monitor that alerts when error rate in logs exceeds 5%.”

• updateMonitor (Write, path) — Update an existing monitor by ID.

  Example: “Update monitor mon_123 to use a 5-minute threshold window.”

• deleteMonitor (Write, path) — Delete a monitor by ID.

  Example: “Delete monitor mon_123.”

• createNotifier (Write, path) — Create a new notifier.

  Example: “Create a Slack notifier for the #alerts channel.”

• updateNotifier (Write, path) — Update an existing notifier by ID.

  Example: “Update notifier ntf_123 to point at a different webhook URL.”

• deleteNotifier (Write, path) — Delete a notifier by ID.

  Example: “Delete notifier ntf_123.”

Service map#

These tools are available when service map is enabled for your organization.

• services-getMap (Read, path) — Get the service dependency graph, including call volume, errors, and approximate latency.

  Example: “Show the service map for the last hour.”

• services-getChanges (Read, path) — Report what changed in the service dependency graph, such as new or missing dependencies and latency regressions.

  Example: “What changed in the service map in the last 6 hours?”

• services-getEdgeHistory (Read, path) — Get time-series history for a service’s dependency edges.

  Example: “Show edge history from api-gateway to checkout over the last day.”

Feedback#

• sendFeedback (Write, path) — Send feedback to the Axiom team about MCP tool descriptions, results, errors, or missing capabilities.

  Example: “Report that the queryDataset time-range description is unclear.”

Prompts#

Axiom MCP Server also ships prebuilt prompts for clients that support MCP prompts. Prompts aren’t available with the local setup.

• explore-unknown-dataset (Prompt, path) — Explore an unknown dataset to understand its structure, content, and potential use cases.

  Example: “Explore the otel-demo dataset and summarize what it’s for.”

• detect-anomalies-in-events (Prompt, path) — Detect anomalies using statistical analysis and pattern recognition across a dataset.

  Example: “Look for anomalies in logs over the last 6 hours.”

• correlate-events-across-datasets (Prompt, path) — Find patterns and correlations between events across multiple datasets.

  Example: “Correlate errors in logs with deploys in cicd for the last day.”

• data-quality-investigation (Prompt, path) — Investigate data quality issues such as missing data, inconsistencies, and collection problems.

  Example: “Check logs for gaps or schema inconsistencies in the last 24 hours.”

• establish-performance-baseline (Prompt, path) — Establish performance baselines for a dataset to support monitoring and anomaly detection.

  Example: “Establish a latency baseline for api over the last 7 days.”

• monitor-health-analysis (Prompt, path) — Analyze monitor health, alert patterns, and effectiveness.

  Example: “Analyze which monitors flap the most this week.”

• investigate-service-topology (Prompt, path) — Walk a system’s service dependency graph to find what talks to what, which dependency is at fault, and when it changed. Available when service map is enabled for your organization.

  Example: “Investigate the topology around api-gateway.”

Axiom plans to support MCP resources in the future.

What’s next#

• Set up Axiom MCP Server
• AI agents