Honeycomb MCP server for AI apps

What your AI apps can do with Honeycomb

Query events and traces
Review boards, SLOs and triggers

Honeycomb MCP tools

34 tools, taken from the publisher's documentation on 29 October 2026. The server decides what it offers and can add or rename tools at any time; after you connect, Toolstile shows the current list.

Tool	What it does
`get_workspace_context`	Get team name, current time and environments with dataset counts
`get_environment`	Get one environment's details, datasets and calculated fields
`get_dataset`	Get metadata and full column schema for one dataset
`get_dataset_columns`	Get column schema with optional sample values or metric names
`run_query`	Run time-series aggregation queries
`get_query_results`	Fetch results of an earlier query run
`find_queries`	Search query history and saved queries by intent
`find_columns`	Find columns and calculated fields by natural-language keywords
`run_bubbleup`	Run BubbleUp to compare a selected subset against the baseline
`get_trace`	Show all spans of a trace as a waterfall
`list_spans`	List span names ranked by count
`get_span_details`	Get attribute summary and top values for spans with a given name
`get_service_map`	Get a snapshot of service-to-service dependencies (Enterprise)
`get_anomaly_service_profiles`	Get anomaly detection service profiles
`search_semconv`	Search the semantic convention registry for attributes
`get_semconv_attribute`	Get full definitions of attributes by exact name
`list_semconv_namespaces`	List top-level namespaces in the team's registry
`list_boards`	List Boards in an environment or return one Board's contents
`create_board`	Create a Board with query, SLO and Markdown panels
`update_board`	Rename a Board and add, remove, update or reorder panels
`get_triggers`	List Triggers or return one Trigger's configuration
`create_trigger`	Create a threshold-based alerting Trigger
`update_trigger`	Modify an existing Trigger
`get_slos`	List SLOs or return one SLO's status and graphs
`create_slo`	Create an SLO and its SLI derived column
`update_slo`	Partially update an existing SLO
`list_recipients`	List email, Slack, PagerDuty and webhook recipients
`create_recipient`	Create a recipient for Triggers and burn alerts
`canvas_agent_invoke`	Message the Canvas agent in a new or existing investigation
`canvas_agent_poll_response`	Poll for the result of an earlier Canvas invoke call
`list_aiconversations`	List recent AI conversation IDs ranked by event count
`get_aiconversation`	Get the full event timeline and summary for one AI conversation
`refinery_docs`	Read Refinery documentation to answer configuration questions
`feedback`	Send feedback about the MCP server to Honeycomb

Suggested access for each AI app

The documentation this list comes from doesn't mark Honeycomb's tools as read-only or destructive. Read each tool's description and decide per app before allowing it.

Coding agent in Claude Code or Cursor. Allow the read tools so the agent can look things up without stopping, and require approval for tools that change shared data.

Chat or research app in Claude Desktop. Allow reading and searching. Allow writes only where you want the app to act for you.

Unattended agent, such as a scheduled Codex job. No one is watching the session, so keep it narrow: allow what the job needs, require approval for changes and block the rest.

Blocked tools are removed from the tool list an app sees, so it won't try them. Because rules apply per app, the same Honeycomb connection can be broad in one app and read-only in another, and the credential stays in Toolstile instead of each app's config file.

How it works with Toolstile

Connect once. Add Honeycomb to a workspace (browser sign-in). Toolstile keeps the credential, so no agent config holds it.
Choose access per app. Give Claude Code every Honeycomb tool, give Cursor only the read tools, or keep Honeycomb away from an app entirely.
Share approved accounts. Teammates use the connection without copying tokens, and you can remove access in one place.
Review activity. See which app called which tool and when, so you can tighten access with evidence.

Add the Honeycomb MCP server to an AI app

Through Toolstile, every AI app uses one gateway entry, whichever servers you connect. In Claude Code:

Shell

claude mcp add --transport http toolstile 'https://toolstile.com/mcp'

Then connect Honeycomb in your workspace and choose which of its tools that app may use. Client setup covers Codex, Cursor and Claude Desktop.

To connect directly instead, add the Honeycomb endpoint to each app separately. In Claude Code:

Shell

claude mcp add --transport http honeycomb 'https://mcp.honeycomb.io/mcp'

In Cursor's mcp.json:

JSON

{
  "mcpServers": {
    "honeycomb": {
      "url": "https://mcp.honeycomb.io/mcp"
    }
  }
}

Some servers accept only clients they have approved, or need a token header. Check the Honeycomb MCP documentation for what your client needs.

Before you connect

A Honeycomb account. Sign-in uses your Honeycomb permissions.

Server details

Detail	Value
Publisher	Honeycomb
Endpoint	`https://mcp.honeycomb.io/mcp`
Authentication	Browser sign-in
Categories	Development
Tool list	34 tools, observed 29 October 2026 (docs)

Honeycomb MCP documentation

Honeycomb MCP server for Claude, Codex and Cursor