Honeycomb

Honeycomb MCP Connector for Claude

A+

Automate observability via Honeycomb — manage datasets, queries, and markers directly from any AI agent.

12 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect your Honeycomb.io observability platform to any AI agent and take full control of your telemetry data, query specifications, and incident markers through natural conversation.

What you can do

  • Dataset Oversight — List all event sources, retrieve detailed metadata, and monitor last access times for your datasets.
  • Query Management — Define new query specifications and execute them to retrieve granular performance insights.
  • Marker Automation — Create timeline annotations (e.g., for deployments or outages) to contextualize your data visualization.
  • Schema Insights — List and inspect columns within specific datasets to understand your event structure.
  • Team Collaboration — Access shared boards and retrieve information about your Honeycomb team configuration.
  • Incident Analysis — Use AI to run complex queries and retrieve results for rapid troubleshooting and RCA.

How it works

  1. Subscribe to this server
  2. Enter your Honeycomb API Key and select your Region (US or EU)
  3. Start querying your telemetry from Claude, Cursor, or any MCP-compatible client

No more manual configuration of complex query DSLs for simple status checks. Your AI assistant acts as a dedicated Observability Engineer or SRE Analyst.

Who is this for?

  • SREs & DevOps — instantly retrieve query results and create markers during incident response.
  • Software Engineers — monitor the performance of new deployments by inspecting specific dataset trends.
  • Platform Leads — maintain a real-time overview of dataset usage and board sharing across the organization.
telemetryhigh-cardinalityincident-responsedata-queryingreal-time-insights

12 tools expose this connector's capabilities to your AI agent.

create_marker

Pass details as a JSON string in "body_json" (requires message). Use "__all__" for team-wide markers. Create a new marker (e.g., deploy, maintenance) on a dataset timeline

create_query_specification

Pass the specification as a JSON string in "query_json". Returns a query ID for execution. Create a new query specification for a dataset

run_query

Poll for results using "get_query_result" with the returned result ID. Execute a query specification and return a result ID

get_dataset_details

Get metadata for a specific dataset

get_team_details

Retrieve information about the Honeycomb team

get_query_result

Retrieve the results of an executed query

list_honeycomb_boards

List all boards (dashboards) shared with the team

list_dataset_columns

List all columns (fields) defined in a specific dataset

list_datasets

Use this to find the "slug" required for markers and queries. List all datasets in your Honeycomb team

list_markers

List markers (annotations) for a dataset

list_queries

List query specifications for a specific dataset

list_triggers

List triggers (alerts) defined for a dataset

See how to talk to your AI agent using Honeycomb.

List all datasets and find one related to 'payment-gateway'.

I've retrieved your datasets. I found 'gateway-prod' (Slug: gateway-prod) and 'gateway-staging'. Which dataset's schema would you like to inspect?

Create a marker on all datasets: 'Deploy v2.4.0 started'.

Marker created! I've successfully added the annotation 'Deploy v2.4.0 started' to all datasets using the environment-wide slug. You should see this marker appear on your graphs in a few moments.

Execute query ID 'q_99283' for the 'main-api' dataset.

Query execution triggered! Your result ID is 'res_552'. I'll wait a few moments for the data to process. Should I retrieve the final results for you now?

Log in to Honeycomb, go to **Team Settings**, and navigate to the **API Keys** section. You will be able to generate and copy your Team API Key from there. Ensure you also note your account's region.

Related Connectors