Grafana-Loki MCP Server

Grafana-Loki MCP Server

Enables querying and formatting Loki logs from Grafana via the Model Context Protocol. It supports LogQL queries, label retrieval, and provides results in text, JSON, or markdown formats.

Category
访问服务器

README

Grafana-Loki MCP Server

Test PyPI version codecov License: MIT

A FastMCP server that allows querying Loki logs from Grafana.

MCP Server Settings

{
  "mcpServers": {
    "loki": {
      "command": "uvx",
      "args": [
        "grafana-loki-mcp",
        "-u",
        "GRAFANA_URL",
        "-k",
        "GRAFANA_API_KEY"
      ]
    }
  }
}
  • GRAFANA_URL: URL of your Grafana instance
  • GRAFANA_API_KEY: Grafana API key with appropriate permissions

Features

  • Query Loki logs through Grafana API
  • Get Loki labels and label values
  • Format query results in different formats (text, JSON, markdown)
  • Support for both stdio and SSE transport protocols

Requirements

  • Python 3.10+
  • FastMCP
  • Requests

Installation

Using pip

pip install grafana-loki-mcp

Development Setup

  1. Clone this repository
  2. Install dependencies using uv:
# Install uv
pip install uv

# Create and activate virtual environment
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
uv pip install -e ".[dev]"

Usage

Environment Variables

Set the following environment variables:

  • GRAFANA_URL: URL of your Grafana instance
  • GRAFANA_API_KEY: Grafana API key with appropriate permissions

Command Line Arguments

You can also provide these values as command line arguments:

grafana-loki-mcp -u https://your-grafana-instance.com -k your-api-key

Additional options:

  • --transport: Transport protocol to use (stdio or sse, default: stdio)

Running the Server

# Using environment variables
export GRAFANA_URL=https://your-grafana-instance.com
export GRAFANA_API_KEY=your-api-key
grafana-loki-mcp

# Using command line arguments
grafana-loki-mcp -u https://your-grafana-instance.com -k your-api-key

# Using SSE transport
grafana-loki-mcp --transport sse

Development

Testing

Run the test suite:

pytest

Run with coverage:

pytest --cov=. --cov-report=term

Linting and Formatting

# Run ruff linter
ruff check .

# Run black formatter
black .

# Run type checking
mypy .

Available Tools

query_loki

Query Loki logs through Grafana.

Parameters:

  • query: Loki query string
  • start: Start time (ISO format, Unix timestamp, or Grafana-style relative time like 'now-1h', default: 1 hour ago)
  • end: End time (ISO format, Unix timestamp, or Grafana-style relative time like 'now', default: now)
  • limit: Maximum number of log lines to return (default: 100)
  • direction: Query direction ('forward' or 'backward', default: 'backward')
  • max_per_line: Maximum characters per log line (0 for unlimited, default: 100)

get_loki_labels

Get all label names from Loki.

get_loki_label_values

Get values for a specific label from Loki.

Parameters:

  • label: Label name

format_loki_results

Format Loki query results in a more readable format.

Parameters:

  • results: Loki query results from query_loki
  • format_type: Output format ('text', 'json', or 'markdown', default: 'text')
  • max_per_line: Maximum characters per log line (0 for unlimited, default: 0)

Example Usage

# Example client code
from mcp.client import Client

async with Client() as client:
    # Query Loki logs with max_per_line limit
    results = await client.call_tool(
        "query_loki",
        {
            "query": '{app="my-app"} |= "error"',
            "limit": 50,
            "max_per_line": 100,  # Limit log lines to 100 characters
            "start": "now-6h",    # Grafana-style relative time: 6 hours ago
            "end": "now"          # Current time
        }
    )

    # Format the results
    formatted = await client.call_tool(
        "format_loki_results",
        {
            "results": results,
            "format_type": "markdown",
            "max_per_line": 100  # Can also limit at formatting time
        }
    )

    print(formatted)

License

This project is licensed under the MIT License - see the LICENSE file for details.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选