Skip to main content
Glama

Server Details

Query Telemetry Machine observability data and instrument your apps with OpenTelemetry.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP · MCP 2025-11-25
URL
ChatGPT
–
Claude
Telemetry Machine

TDQS

B3.4/5.0

Scored across 28 tools

Disambiguation4/5

Each tool maps to a distinct resource and action (datasets, dashboards, widgets, SLOs, saved views, annotations, queries, guides), and the edit_* vs get_* split is clear. The main wrinkle is API keys: create_ingest_key creates one while edit_api_keys updates/deletes them, so an agent could momentarily conflate the two, and get_widgets bundles several operations (list/get/run/preview/schema) that could be mistaken for run_query.

Naming Consistency4/5

Names are consistently snake_case with a leading verb (create_, edit_, get_, list_, note_, post_, record_, run_), which makes the pattern predictable. Minor deviations: 'feedback' is a bare noun, and the same resource is called 'ingest_key' in create_ingest_key but 'api_keys' in edit_api_keys/get_api_keys, plus singular/plural resource names vary (create_dataset vs list_datasets).

Tool Count3/5

At 28 tools the surface is heavy, sitting just above the 25-tool threshold, and several edit_* tools each bundle create/update/delete. However, the domain genuinely spans many resources (datasets, keys, dashboards, widgets, views, SLOs, annotations, queries, guides, orgs), so most tools do earn their place rather than being redundant.

Completeness4/5

Core lifecycle coverage is strong: datasets, ingest keys, dashboards, widgets, saved views, SLOs and annotations all have read plus create/update/delete paths, and query/schema/semconv/onboarding guides round out the workflow. Gaps are modest—no delete_dataset, no dataset metadata update beyond notes, and no way to edit or delete an annotation—but agents can work around most of these.

Available Tools

28 tools
create_datasetCreate datasetCInspect

Create a dataset for one signal.

Inputs: name (required), signal_type (required).

create_ingest_keyCreate ingest keyAInspect

Create an API key that can only send data to the given datasets. The result holds the key, shown only this once, and the endpoint and headers to send with it.

Inputs: name (required), datasets (required).

edit_api_keysEdit API keysA
Destructive
Inspect

Update or delete an API key. update body: {"name"?, "description"?, "add_datasets"?: [{"dataset_public_id"}], "remove_datasets"?: [dataset_public_id]}. Needs an owner or admin.

Inputs: action (required), id, body.

edit_dashboardsEdit dashboardsA
Destructive
Inspect

Create, update or delete a dashboard, and add, lay out, reorder or remove its tiles, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md). Bodies: create {"name", "description"?, "widget_ids"?}; update {"name"?, "description"?, "filters"?: {"relative_time_range_seconds"?, "date_from"?, "date_to"?, "variables"?: [{"name", "field", "label"?, "value"?}]}} (filters replaces the whole object; a variable filters every tile whose dataset has its field, like the "filters" of a widget query); add_tile {"widget_id"}; update_tile {"layout_options"?: {"x", "y", "w", "h"}, "overrides"?: {"links": [{"field", "dashboard_id", "variable"}]}} on a 12-column grid (tiles without a layout fill two columns, 6 by 4, in order; a link opens that dashboard, which must define the variable, with the variable set to the field in the clicked row); reorder_tiles {"tile_ids"}. Deleting a dashboard keeps its widgets.

Inputs: action (required), id, body, tile_id.

edit_saved_viewsEdit saved viewsA
Destructive
Inspect

Create, update or delete one of the user's saved explorer views of a traces or logs dataset. update changes only what you give.

Inputs: action (required), id, name, since, +5 more.

edit_slosEdit SLOsB
Destructive
Inspect

Create, update or delete an SLO, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md). Body: {"name", "definition", "description"?, "owner"?}, the definition as its JSON Schema describes.

Inputs: action (required), id, body.

edit_widgetsEdit widgetsB
Destructive
Inspect

Create, update or delete a widget, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md). The body of create and update is a widget, as its JSON Schema describes.

Inputs: action (required), id, body.

feedbackFeedbackAInspect

Send feedback about these tools to the Telemetry Machine team: something missing, wrong or confusing.

Inputs: message (required).

get_annotationsGet annotationsA
Read-only
Inspect

List annotations, newest first: the deploys, changes and comments the team or its CI posted. A window has an end; a moment does not.

Inputs: type, limit, since, until, +1 more.

get_api_keysGet API keysB
Read-only
Inspect

List the organization's API keys, or get one. Secrets are never shown.

Inputs: action (required), id, body, limit, +2 more.

get_dashboardsGet dashboardsA
Read-only
Inspect

List dashboards, or get one with its tiles, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md).

Inputs: action (required), id, body, limit, +2 more.

get_field_valuesField valuesC
Read-only
Inspect

List the distinct values of a field.

Inputs: field (required), dataset (required).

get_instrumentation_guideInstrumentation guideA
Read-only
Inspect

Get the steps to onboard another service once the user already sends some data to Telemetry Machine. For when the user wants to send more data, or instrument or add another service.

Inputs: no named parameters.

get_onboarding_guideOnboarding guideA
Read-only
Inspect

Get the steps that get the user's app sending telemetry to Telemetry Machine. For when the user wants to set up, connect, get started with or continue onboarding Telemetry Machine.

Inputs: no named parameters.

get_organizationsGet organizationsB
Read-only
Inspect

List the organizations this connection reaches, or the members or pending invitations of one.

Inputs: action (required), limit, offset.

get_query_resultQuery resultA
Read-only
Inspect

Get the rows of a slow query by its query ID, or learn that it is still running.

Inputs: query_id (required).

get_saved_viewsGet saved viewsA
Read-only
Inspect

List the user's saved explorer views, or get one with a link that opens it.

Inputs: action (required), id, dataset.

get_schemaGet schemaA
Read-only
Inspect

Get a dataset's fields, their types and the notes on them. Pass search to narrow the fields by words in their names or notes.

Inputs: dataset (required), search.

get_semconvGet semantic conventionsA
Read-onlyIdempotent
Inspect

Look up the OpenTelemetry semantic conventions: the standard names of attributes and metrics, with their meanings. For choosing names when adding instrumentation.

Inputs: action (required), kind, limit, names, +2 more.

get_slosGet SLOsB
Read-only
Inspect

List SLOs, each with its status and error budget left, or get one, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md). events gives an SLO's recent state changes; schema gives the JSON Schema of a definition.

Inputs: action (required), id, body, limit, +2 more.

get_widgetsGet widgetsA
Read-only
Inspect

List, get or run saved widgets, through the Telemetry Machine REST API (https://telemetrymachine.com/docs/api.md). query runs a saved widget (body: {"time_range"?: {"last": "1h"}, "filters"?: [...]}, the filters ANDed with its own); preview runs an unsaved one (body: {"definition": ...}); schema gives the JSON Schema of a widget.

Inputs: action (required), id, body, limit, +2 more.

list_datasetsList datasetsA
Read-only
Inspect

List every dataset with its type (traces, logs, metrics, events), description, and event count over the past 24 hours.

Inputs: no named parameters.

note_datasetNote datasetA
Idempotent
Inspect

Save a one-line description of what a dataset holds. It fills in a missing description; one already there stays.

Inputs: dataset (required), description (required).

note_fieldNote fieldA
Idempotent
Inspect

Save a field's description, unit and labels in the dataset's schema, or remove them. Only the parts you give change; a person's note is never replaced.

Inputs: field (required), dataset (required), unit, clear, +3 more.

post_annotationPost an annotationAInspect

Record something that explains the data on the timeline: a change you made (a config edit, a flag flip), a deploy, or a comment. It is a record and runs nothing. Put what it concerns in attributes with OpenTelemetry names, such as service.name. A deploy needs vcs.ref.head.revision or service.name. Posting the same key again with the same fields is a safe retry.

Inputs: type (required), message (required), key, dataset, +3 more.

record_onboarding_stateRecord onboarding stateB
Idempotent
Inspect

Record the step of Telemetry Machine's onboarding that is starting, to track the user's progress through it.

Inputs: state (required).

run_queryRun queryA
Read-only
Inspect

Query one dataset with Apache DataFusion SQL (https://datafusion.apache.org/user-guide/sql/). Use raw_sql for joins, subqueries, CTEs and window functions. A slow query returns a query ID in place of rows.

Inputs: dataset (required), description (required), limit, since, +6 more.

run_surfaceRun surfaceB
Read-only
Inspect

Find the field values most over-represented in a selection of spans or logs compared with the rest. Kubernetes pod and node metrics compare too: each event gets what its pod and node reported in the minute it happened, such as the pod's CPU or memory against its limit.

Inputs: slo, since, until, where, +4 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      C
      maintenance
      Minimal agent telemetry store with OTLP/HTTP ingestion to Postgres and MCP streamable-http tools for querying logs, traces, and metrics.
      -
    • A
      license
      Not graded
      quality
      B
      maintenance
      Self-hosted OpenTelemetry backend in one Go binary. Its MCP server exposes tools for service health, latency, topology, trace inspection, log search, read-only SQL over traces, logs and metrics, and dashboards.
      129
      Apache 2.0
    • A
      license
      A
      quality
      C
      maintenance
      Enables AI agents to query OriginChain's telemetry agent for log patterns, semantic incident analysis, distributed traces, service graphs, and read-only SQL queries over telemetry data.
      10
      4 npm
      MIT
    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources