Loki MCP Server

What does “Loki” MCP Server do?

The Loki MCP Server is a Go-based implementation of the Model Context Protocol (MCP) designed to integrate with Grafana Loki, a log aggregation system. It serves as a bridge between AI assistants and external log data sources, enabling the AI to query and interact with log streams stored in Loki. By exposing Loki’s querying capabilities via the MCP protocol, developers and AI clients can enhance their workflows—such as searching, filtering, and analyzing logs—directly through standardized LLM-driven interfaces. This empowers tasks like real-time log investigation, troubleshooting, and dashboard creation, providing seamless access to operational data for improved observability and automation.

List of Prompts

No prompt templates are documented in the repository.

List of Resources

No explicit MCP resources are described in the repository.

List of Tools

• loki_query
  Allows querying Grafana Loki log data.
  • Required parameters:
    • query: LogQL query string
  • Optional parameters:
    • url: Loki server URL (default from LOKI_URL env or http://localhost:3100)
    • start: Start time for the query (default: 1 hour ago)
    • end: End time for the query (default: now)
    • limit: Max number of entries to return (default: 100)

Use Cases of this MCP Server

• Log Data Exploration
  Developers or AI agents can programmatically query and analyze log data stored in Grafana Loki, assisting with troubleshooting and incident response.
• Automated Log Monitoring
  Enables LLM-driven monitoring workflows, where AI assistants can scan logs for anomalies, error patterns, or specific events.
• Operational Dashboards
  Facilitates dynamic dashboard creation by fetching log data required for visualization or reporting tools.
• Root Cause Analysis
  Empowers AI to sift through large volumes of logs, pinpointing the origin of issues using flexible LogQL queries.

How to set it up

Windsurf

1. Install Go 1.16 or higher.

2. Build the server:
   go build -o loki-mcp-server ./cmd/server

3. Edit your Windsurf configuration to add the MCP server.

4. Add the Loki MCP server with a JSON snippet (adapt as needed):

   {
     "mcpServers": {
       "loki-mcp": {
         "command": "./loki-mcp-server",
         "args": []
       }
     }
   }
   

5. Save the configuration and restart Windsurf.

6. Verify the server is running and accessible.

Securing API Keys (Environment Variables Example):

{
  "mcpServers": {
    "loki-mcp": {
      "command": "./loki-mcp-server",
      "env": {
        "LOKI_URL": "https://your-loki-server.example"
      }
    }
  }
}

Claude

1. Install Go 1.16 or higher.

2. Build the server as above.

3. Open Claude’s MCP configuration file.

4. Add the Loki MCP server:

   {
     "mcpServers": {
       "loki-mcp": {
         "command": "./loki-mcp-server",
         "args": []
       }
     }
   }
   

5. Save/restart Claude.

6. Confirm setup is working.

Securing API Keys:

{
  "mcpServers": {
    "loki-mcp": {
      "command": "./loki-mcp-server",
      "env": {
        "LOKI_URL": "https://your-loki-server.example"
      }
    }
  }
}

Cursor

1. Ensure Go 1.16+ is installed.

2. Build Loki MCP server.

3. Edit Cursor’s configuration.

4. Add Loki MCP server entry:

   {
     "mcpServers": {
       "loki-mcp": {
         "command": "./loki-mcp-server",
         "args": []
       }
     }
   }
   

5. Save and restart Cursor.

6. Verify integration.

Using Environment Variables:

{
  "mcpServers": {
    "loki-mcp": {
      "command": "./loki-mcp-server",
      "env": {
        "LOKI_URL": "https://your-loki-server.example"
      }
    }
  }
}

Cline

1. Install Go >=1.16.

2. Build with:
   go build -o loki-mcp-server ./cmd/server

3. Locate Cline’s MCP server config.

4. Add Loki MCP server:

   {
     "mcpServers": {
       "loki-mcp": {
         "command": "./loki-mcp-server",
         "args": []
       }
     }
   }
   

5. Save and restart Cline.

6. Test the setup.

Secure API Keys via env:

{
  "mcpServers": {
    "loki-mcp": {
      "command": "./loki-mcp-server",
      "env": {
        "LOKI_URL": "https://your-loki-server.example"
      }
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "loki-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change "loki-mcp" to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Based on the above tables, Loki MCP Server offers a clear overview and a functional tool for querying logs, but lacks documented prompts, resources, and advanced MCP features like sampling or roots. The documentation is minimal, and setup is developer-oriented.

Our opinion

The Loki MCP Server is focused and functional for integrating LLMs with Grafana Loki log querying, but it is minimalistic and lacks breadth in MCP features and documentation. For a score, it would receive a 4/10: it works for its main purpose, but is not a feature-complete, polished, or highly documented MCP server.

MCP Score