Language: 日本語English

honeycomb Plugin

Author
Honeycomb
Category
Monitoring
Topics
Monitoring & Observability · AI Agents & AI App Development · Claude Code Customization & Workflow
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

This is a plugin named "honeycomb" that adds skills, agents, and workflows for Honeycomb observability to AI coding agents, designed to complement the Honeycomb MCP server. According to the README, it includes 9 skills covering observability fundamentals, query patterns, metrics queries, production investigation, SLOs & triggers, OpenTelemetry instrumentation, OpenTelemetry migration, Beeline migration, and board creation. It also ships 2 agents — honeycomb-investigator for autonomous multi-step production debugging and instrumentation-advisor for codebase-to-Honeycomb gap analysis — plus a /honeycomb-setup command for interactive MCP server configuration. Full plugin installation is supported in Claude Code, OpenAI Codex, Cursor, Augment (Auggie CLI), and GitHub Copilot CLI; VS Code Copilot, OpenAI Codex CLI, and Cline require manual skills + MCP config, while Windsurf, Amazon Q Developer, Continue, and Copilot Coding Agent can only connect to the MCP server directly.

Overview

According to the Honeycomb homepage, Honeycomb is presented as an "AI-Ready Observability Platform" offering distributed tracing, log analytics, custom metrics, frontend observability, and LLM/AI agent observability, built on a purpose-built columnar data store. Listed platform features include SLOs, Service Map, BubbleUp, Query Builder, the Canvas AI copilot, and an MCP server for connecting AI agents to observability data.

What you can do with honeycomb

  • Install the Honeycomb skills, agents, and the /honeycomb-setup command as a full plugin in Claude Code, Cursor, OpenAI Codex, Augment (Auggie CLI), or GitHub Copilot CLI
  • Use the honeycomb-investigator agent for autonomous multi-step production debugging
  • Use the instrumentation-advisor agent for codebase-to-Honeycomb gap analysis
  • Draw on 9 skills covering query patterns, metrics queries, production investigation, SLOs & triggers, OpenTelemetry instrumentation/migration, Beeline migration, and board creation
  • Manually set up skills plus an MCP config in VS Code Copilot, OpenAI Codex CLI, or Cline, or connect directly to the MCP server from Windsurf, Amazon Q Developer, Continue, or Copilot Coding Agent
  • Run the included structural tests (make test) to validate plugin layout without needing API keys

Sources

Original description (English)

Skills, agents, and workflows for Honeycomb observability — query patterns, production investigations, SLOs, OpenTelemetry instrumentation, and Beeline migration. Designed to complement the Honeycomb MCP server.

History of honeycomb

Back to list