Frequently Asked Questions about Honeycomb MCP

Q: What is Honeycomb MCP? A: Honeycomb MCP (Model Context Protocol) server allows Large Language Models (LLMs) to directly query and analyze your Honeycomb observability data, enabling AI-powered insights.

Q: What are the requirements for using Honeycomb MCP? A: You need Node.js 18+, a Honeycomb API key with full permissions, and a Honeycomb Enterprise account.

Q: How do I install Honeycomb MCP? A: Use the following commands: bash pnpm install pnpm run build

The build artifact will be in the /build folder.

Q: How do I configure Honeycomb MCP? A: You need to provide Honeycomb API keys via environment variables in your MCP config. Example:

{ “mcpServers”: { “honeycomb”: { “command”: “node”, “args”: [ “/fully/qualified/path/to/honeycomb-mcp/build/index.mjs” ], “env”: { “HONEYCOMB_API_KEY”: “your_api_key” } } } }

Q: What clients are compatible with Honeycomb MCP? A: Honeycomb MCP has been tested with Claude Desktop, Claude Code, Cursor, Windsurf, and Goose. It will likely work with other clients as well.

Q: What kind of queries can I run with Honeycomb MCP? A: You can run analytics queries with support for multiple calculation types (COUNT, AVG, P95, etc.), breakdowns, filters, and time-based analysis. You can also monitor SLOs and their status, analyze columns and data patterns, and view and analyze Triggers.

Q: How do I access Honeycomb datasets using Honeycomb MCP? A: Access Honeycomb datasets using URIs in the format honeycomb://{environment}/{dataset}. For example: honeycomb://production/api-requests

Q: What are some example queries I can use with Claude and Honeycomb MCP? A: Here are some example questions you can ask Claude:

• “What datasets are available in the production environment?”
• “Show me the P95 latency for the API service over the last hour”
• “What’s the error rate broken down by service name?”
• “Are there any SLOs close to breaching their budget?”
• “Show me all active triggers in the staging environment”
• “What columns are available in the production API dataset?”

Q: How is Honeycomb MCP optimized for LLMs? A: All tool responses are optimized to reduce context window usage while maintaining essential information, ensuring that responses are concise but complete, allowing LLMs to process more data within context limitations.

Q: Can I configure caching for Honeycomb MCP? A: Yes, you can configure caching using environment variables. You can enable/disable caching, set default TTL values, and configure resource-specific TTL values.

Q: I am an EU customer; do I need a special configuration? A: Yes, EU customers must set a HONEYCOMB_API_ENDPOINT configuration to point to the EU instance: HONEYCOMB_API_ENDPOINT=https://api.eu1.honeycomb.io/

Q: How does Honeycomb MCP integrate with the UBOS platform? A: By integrating Honeycomb MCP with UBOS, you can orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model, and create Multi-Agent Systems to automate observability workflows.