generated: '2026-08-24' method: probed source: https://docs.honeycomb.io/integrations/mcp/ and live probes of https://mcp.honeycomb.io/mcp and https://docs.honeycomb.io/mcp provider: Honeycomb providerId: honeycomb-io status: published description: >- Honeycomb operates a first-party hosted Model Context Protocol server at https://mcp.honeycomb.io/mcp (EU: https://mcp.eu1.honeycomb.io/mcp). It is OAuth-protected: an unauthenticated POST returns 401 with a WWW-Authenticate challenge pointing at /.well-known/oauth-protected-resource, which names https://ui.honeycomb.io as the authorization server and mcp:read / mcp:write as the supported scopes. A separate, anonymous documentation MCP server runs on the docs host at https://docs.honeycomb.io/mcp and answers tools/list without credentials. deployment: mode: remote endpoint: https://mcp.honeycomb.io/mcp auth: oauth verified: probed servers: - name: Honeycomb MCP (US) endpoint: https://mcp.honeycomb.io/mcp mode: remote auth: oauth alternate_auth: >- Management API key formatted KEY_ID:SECRET_KEY passed as a Bearer token, for headless agents that cannot complete an interactive OAuth flow. The key needs the "Model Context Protocol" scope (Read, plus Write for mutating tools) and Environments. authorization_server: https://ui.honeycomb.io protected_resource_metadata: https://mcp.honeycomb.io/.well-known/oauth-protected-resource scopes: - mcp:read - mcp:write verified: probed probe: url: https://mcp.honeycomb.io/mcp method: POST body: '{"jsonrpc":"2.0","id":1,"method":"initialize"}' http_status: 401 www_authenticate: Bearer resource_metadata="https://mcp.honeycomb.io/.well-known/oauth-protected-resource" note: >- The 401 is the expected, correct behaviour for an OAuth-protected MCP resource; it proves the endpoint is live and standards-compliant (RFC 9728). The live tool schemas require an authenticated introspection, so the tools below are transcribed from the provider's own published tools reference. - name: Honeycomb MCP (EU) endpoint: https://mcp.eu1.honeycomb.io/mcp mode: remote auth: oauth verified: searched source: https://docs.honeycomb.io/integrations/mcp/ - name: Honeycomb Docs MCP endpoint: https://docs.honeycomb.io/mcp mode: remote auth: none verified: probed scope: documentation search and retrieval over the published docs site only probe: url: https://docs.honeycomb.io/mcp method: POST http_status: 200 server_info: 'Honeycomb Docs 1.0.0, protocolVersion 2024-11-05' tools: - name: search_honeycomb_docs input: [query] read_only: true - name: query_docs_filesystem_honeycomb_docs input: [command] read_only: true note: >- Read-only virtual filesystem over the docs corpus and OpenAPI specs; supports rg, grep, find, tree, ls, cat, head, tail, jq and similar text utilities. - name: submit_feedback input: [path, feedback] read_only: false deprecated_deployments: - name: honeycomb-mcp (self-hosted stdio) mode: local-stdio package: https://github.com/honeycombio/honeycomb-mcp status: deprecated archived: true archived_source: 'GitHub repository archived; last push 2025-08-21' note: >- The repository README states plainly that the self-hosted MCP server is deprecated and directs users to the hosted server. Recorded here so the local-stdio history is not confused with the current remote surface. tools: source: https://docs.honeycomb.io/integrations/mcp/tools/ method: searched note: >- Transcribed from Honeycomb's published MCP tools reference. Input schemas are not reproduced because the live server requires authenticated introspection. catalog: - category: Workspace discovery read: - name: get_workspace_context description: Returns the team name, current time, and environments with slugs and dataset counts. Takes no parameters. rate_limit_per_minute: 200 - name: get_environment description: Returns details for a single environment, including datasets and calculated fields. Up to 100 datasets per call. rate_limit_per_minute: 200 - name: get_dataset description: Returns metadata and the full column schema for a single dataset. Up to 100 columns per call by default. rate_limit_per_minute: 200 - name: get_dataset_columns description: Returns the full column schema for a dataset, with optional sample values; for metrics datasets returns metric names. rate_limit_per_minute: 100 - category: Query and analysis read: - name: run_query description: Runs a time-series aggregation query against a dataset and returns computed results. Supports query math, per-calculation filters, formulas, breakdowns, calculated fields and relational trace prefixes. rate_limit_per_minute: 100 - name: get_query_results description: Retrieves results and metadata from a previously executed query run by URL, run primary key, or query ID. rate_limit_per_minute: 150 - name: find_queries description: Searches query history and saved queries by intent and returns matching queries with their run primary keys. rate_limit_per_minute: 150 - name: find_columns description: Searches columns and calculated fields by intent across one or all datasets in an environment. rate_limit_per_minute: 200 - name: run_bubbleup description: Runs a BubbleUp analysis on an existing query result to identify what makes a selected subset different from the baseline. rate_limit_per_minute: 50 - category: Traces and spans read: - name: get_trace description: Retrieves all spans for a trace ID and renders them as a waterfall. rate_limit_per_minute: 100 - name: list_spans description: Lists span names in trace data ranked by count, with root frequency and source dataset. rate_limit_per_minute: 50 - name: get_span_details description: Returns a summary of attributes and common values observed on spans with a specific name. rate_limit_per_minute: 50 - category: Service map and anomalies read: - name: get_service_map description: Returns a snapshot of service-to-service call dependencies for a time range. Enterprise tier. rate_limit_per_minute: 10 tier: Enterprise - name: get_anomaly_service_profiles description: Returns anomaly detection service profiles for the current team. Requires anomaly detection to be enabled. rate_limit_per_minute: 150 - category: Semantic conventions read: - name: search_semconv description: Searches the OpenTelemetry semantic convention registry for attributes matching a query. rate_limit_per_minute: 300 - name: get_semconv_attribute description: Returns full definitions of one or more semantic convention attributes by exact name. rate_limit_per_minute: 300 - name: list_semconv_namespaces description: Lists top-level semantic convention namespaces available in the team's registry. rate_limit_per_minute: 300 - category: Boards read: - name: list_boards description: Lists Boards in an environment, or returns the full contents of a single Board by ID. rate_limit_per_minute: 200 write: - name: create_board description: Creates a Board with query panels, SLO panels and Markdown text panels, with optional preset filters. scope: mcp:write rate_limit_per_minute: 50 - name: update_board description: Updates an existing Board - rename, add/remove/update/reorder panels, replace preset filters and tags. scope: mcp:write rate_limit_per_minute: 50 - category: Triggers read: - name: get_triggers description: Lists Triggers for the team, or returns detailed configuration for a single Trigger including recipients. rate_limit_per_minute: 150 write: - name: create_trigger description: Creates a Trigger that fires alerts when query results cross a threshold. scope: mcp:write rate_limit_per_minute: 20 - name: update_trigger description: Updates an existing Trigger. scope: mcp:write rate_limit_per_minute: 30 - category: SLOs read: - name: get_slos description: Lists SLOs for the team, or returns detailed status and graphs for a single SLO. rate_limit_per_minute: 100 write: - name: create_slo description: Creates an SLO with an auto-created SLI derived column; validates the expression before creating. scope: mcp:write rate_limit_per_minute: 20 - name: update_slo description: Updates an existing SLO with partial-update semantics. Dataset associations are fixed at creation. scope: mcp:write rate_limit_per_minute: 30 - category: Notification recipients read: - name: list_recipients description: Lists pre-registered notification recipients for the team - email, Slack, PagerDuty and webhooks - with IDs. rate_limit_per_minute: 200 write: - name: create_recipient description: Creates a notification recipient that can be attached to Triggers and SLO burn alerts. scope: mcp:write rate_limit_per_minute: 20 - category: Canvas investigations read: - name: canvas_agent_poll_response description: Polls for the result of a previously issued canvas_agent_invoke call. rate_limit_per_minute: 300 write: - name: canvas_agent_invoke description: Sends a message to the Canvas agent, routed into an existing investigation or creating a new one. scope: mcp:write rate_limit_per_minute: 20 - category: AI conversation analysis read: - name: list_aiconversations description: Lists recent AI agent conversations (gen_ai.conversation.id) in an environment with agents, services, event counts, errors and token usage. rate_limit_per_minute: 50 - name: get_aiconversation description: Returns the full event timeline for a single AI conversation, including every LLM call, tool call and related agent event. rate_limit_per_minute: 50 - category: Other read: - name: refinery_docs description: Reads from Honeycomb Refinery documentation so an agent can answer Refinery configuration questions inline. rate_limit_per_minute: 200 write: - name: feedback description: Submits feedback about the Honeycomb MCP server experience to the maintaining team. scope: mcp:write rate_limit_per_minute: 100 rate_limits: default_per_minute: 50 note: Rate-limit responses include the time at which the caller may retry. Limits are documented as subject to change. agent_skills: repository: https://github.com/honeycombio/agent-skill docs: https://docs.honeycomb.io/integrations/agent-skills/ note: >- Honeycomb ships an official skills/agents/hooks plugin that wires the MCP server into Claude Code, Cursor, Augment and GitHub Copilot CLI. See skills/_index.yml.