1. Docs
2. Integrations
3. Model Context Protocol

Honeycomb Model Context Protocol (MCP)

Beta

Explore the Honeycomb MCP server’s core features and capabilities.

What is Honeycomb MCP? 

Honeycomb MCP is an implementation of the Model Context Protocol (MCP), an open standard that enables seamless communication between AI assistants and external data sources. Honeycomb’s hosted MCP server lets agents directly query, analyze, and visualize your observability data—including traces, triggers, and SLOs—using natural language.

What Can You Do With Honeycomb MCP? 

Honeycomb MCP gives AI agents and assistants access to Honeycomb’s observability data and metadata, so they can:

• Run queries across traces, metrics, and logs
• Check the state of Triggers and SLOs
• Analyze and identify patterns, anomalies, and errors
• Suggest instrumentation improvements or optimizations

Key Features 

Use Honeycomb in Your CLI, IDE, or Other Agent Interfaces 

Use Honeycomb from the tools you already rely on, like your IDE, CLI, or AI assistants. Task agents to:

• Investigate and remediate performance issues or bugs
• Assist in migrating dashboards and alerts
• Help explore and present system data

One-Click Install 

Honeycomb MCP lets teams easily onboard members via OAuth 2.1 for fast, secure setup. No API keys to share. Just configure the MCP endpoint, authorize in your browser, and you are ready to go.

AI-Optimized Tooling 

Honeycomb MCP offers unique tools designed with AI agents in mind. More than just a wrapper around our public API, MCP gives agents access to tools that let them fetch single traces, raw data rows, and more.

Getting Started with the MCP Beta 

To join the beta and connect your agents to Honeycomb:

1. Enable Honeycomb Intelligence
2. Configure your AI assistant
3. Put Honeycomb MCP to work

Enable Honeycomb Intelligence 

To use Honeycomb MCP, your team must be enrolled in Honeycomb Intelligence. Once your team enables Honeycomb Intelligence, you will automatically be granted access to MCP.

Configure Your AI Assistant 

Follow our configuration guide to connect your preferred AI assistant. We support many popular tools, including:

• Cursor
• Claude Code/Desktop
• Visual Studio Code
• Amazon Q
• and many more

Establish a secure connection between your AI assistant and Honeycomb MCP via OAuth or API key, depending on your setup.

Put Honeycomb MCP to Work 

Once connected, you can ask agents to explore data, improve instrumentation, or investigate performance anomalies using natural language.

Need ideas or inspiration? To explore real-world use cases of agents using MCP with Honeycomb, visit Honeycomb MCP Example Use Cases.

Beta Limitations 

During the beta period, please be aware that:

• Tool rate limits apply based on your Honeycomb plan.
• Tool definitions, responses, response formats, and other features may change.

Sharing Feedback 

Your input helps us improve Honeycomb MCP. We are especially interested in feedback about:

• Tool coverage and usefulness
• Performance and response times
• Setup experience and integration complexity
• Feature requests or use case ideas

To provide feedback:

• Join our Pollinators Community Slack and share in the #discuss-mcp channel.
• Use the feedback tool directly from your agent.

Next Steps 

Continue your MCP journey with these focused resources:

• Core Concepts: Get familiar with the core concepts behind Model Context Protocol and how Honeycomb MCP lets AI agents interact with observability data.
• Connecting AI Agents to Honeycomb MCP: Follow step-by-step instructions to connect common agents to Honeycomb MCP.
• Example Use Cases: Explore real-world use cases and tips for working with Honeycomb via MCP.
• Troubleshooting: Find solutions to common configuration issues and learn how to verify that your agent is connected and working correctly.