Who it is for
Engineers on call or debugging production issues who already send telemetry to Honeycomb.
Overview
Honeycomb opens an observability workspace to ChatGPT. Traces can be pulled up and read, queries and BubbleUp analyses can be run against event data, and recorded AI conversations are inspected the same way as any other telemetry.
Beyond reading, the app reaches account objects: boards, service level objectives, triggers, markers, and the recipients who receive notifications when something fires. That covers a fair amount of routine maintenance work that would otherwise mean opening the web interface.
A set of bundled skills handles instrumentation and migration. They cover OpenTelemetry setup, instrumentation for GenAI workloads, migration away from Beeline, metrics queries, board creation, and a walkthrough for investigating a production problem.
Engineers on call or debugging production issues who already send telemetry to Honeycomb.
BubbleUp analysis and SLO management are both reachable from chat, and traces from AI conversations are treated as ordinary telemetry.
Availability
honeycomb.io ships this on both marketplaces: Honeycomb is listed in the Claude connectors directory and in the ChatGPT apps directory, under the same publisher domain.
Only 392 of the 3,150 products in this directory are listed in both marketplaces. One MCP server can serve both surfaces; what differs is each platform's listing and review process.
ChatGPT listing
Honeycomb declares 11 named skills, which puts it in the 6.7% of ChatGPT-listed apps that do.
Claude listing
Honeycomb publishes 13 tools to Claude. That is a mid-sized surface: 54.3% of the 1,253 Claude-listed connectors expose fewer.
The median Claude-listed connector filed under Data & Analytics publishes 12, so this one runs one tool above its bucket's median and ranks 84th of 198 by tool count.
It falls in the 11-20 tools band, which holds 22.5% of the Claude directory.
Honeycomb publishes a remote MCP endpoint. Claude connects to it over HTTP; there is nothing to install locally.
Every connector in the Claude directory is remote, all 1,253 of them, so transport is not a differentiator there.
https://mcp.honeycomb.io/mcp Honeycomb carries partner verification in the Claude directory, which 474 of 1,253 listings do (37.8%). That is a review step past self-publishing, not an Anthropic-built server.
In the publisher's words
Query and explore observability data from OpenTelemetry or dozens of other sources in your honeycomb.io account.
Quoted from the connector's own Claude directory listing as captured in August 2026. Node8 did not write it and does not vouch for it.
Nearby
Governed SQL querying across warehouses with a BI layer attached
Claude
Read-only SQL against ClickHouse Cloud services and schemas
Claude
Query Coralogix logs, metrics, and traces conversationally
Claude · ChatGPT
Browse, query and run dbt projects and the semantic layer
Claude
Table search, field exploration and query building in Metabase
Claude
Query a Rill project's metrics views from a conversation
Claude
This is an independent catalogue entry compiled by Node8 from Honeycomb's public marketplace listings as they stood in August 2026. Listing facts come from the marketplaces; the comparisons against the rest of the directory and any live-endpoint measurements are Node8's. Node8 is not affiliated with honeycomb.io, and inclusion here is not an endorsement.
Node8 builds MCP servers end to end: deciding which tools are worth exposing, the authentication and permission model, the review submissions, and the onboarding that gets them used. One server serves ChatGPT, Claude, and Microsoft Copilot.