Solana Metrics MCP Server
Analyzes Solana blockchain metrics from InfluxDB and generates categorized Grafana dashboards for monitoring consensus, network, banking, accounts, RPC, performance, and MEV metrics.
README
Solana Metrics MCP Server
This MCP server analyzes Solana metrics from InfluxDB and generates Grafana dashboards. It's designed to work with the sol_metrics database and provides intelligent categorization and analysis of blockchain metrics.
Features
- 🔍 Metric Discovery: Lists all available metrics from the sol_metrics InfluxDB database
- 🏷️ Intelligent Categorization: Automatically categorizes metrics into logical groups (Consensus, Network, Banking, Accounts, RPC, Performance, Jito/MEV)
- 📊 Metric Analysis: Provides detailed explanations of what each metric measures and why it's useful
- 📁 Auto-Dashboard Export: Creates and saves importable Grafana dashboard JSON files to
grafana/folder - 🔎 Code Search: Helps locate metric definitions in the Solana Rust codebase
- 🔄 Dual InfluxDB Support: Compatible with both InfluxDB v1 and v2
- 🛠️ VS Code Integration: Works seamlessly as an MCP server in Visual Studio Code
Installation
npm install
npm run build
Configuration
Set the following environment variables to connect to your InfluxDB instance:
export INFLUX_URL="http://your-influxdb-server:8086"
export INFLUX_TOKEN="your_token"
export INFLUX_ORG="your_org"
export INFLUX_BUCKET="sol_metrics"
export INFLUX_VERSION="v1" # or "v2" for InfluxDB v2
Dashboard Export
When you generate dashboards using the generate_dashboard tool, the server automatically:
- Saves JSON files to the
grafana/folder in your project directory - Creates timestamped backups for version history
- Provides ready-to-import Grafana dashboard configurations
Generated dashboard files:
consensus-dashboard.json- Epoch rewards, slot confirmation, validator votingnetwork-dashboard.json- Gossip, cluster info, retransmit metricsbanking-dashboard.json- Transaction processing, prioritization feesaccounts-dashboard.json- Account database, cache, hashing metricsrpc-dashboard.json- RPC service and subscription metricsperformance-dashboard.json- CPU, memory, disk usage metricsjito-mev-dashboard.json- Block engine, bundle processing, MEV relayer metrics
Usage with Claude Desktop
Add this server to your Claude Desktop configuration:
{
"mcpServers": {
"solana-metrics": {
"command": "node",
"args": ["/absolute/path/to/solana-metrics-mcp-server/build/index.js"]
}
}
}
Available Tools
1. list_metrics
Lists all available metrics from the sol_metrics database.
2. analyze_metrics
Analyzes and categorizes metrics with detailed explanations.
- Parameters:
category(optional): Filter by category (Consensus, Network, Banking, Accounts, RPC, Performance, Jito/MEV, Other, All)
3. generate_dashboard
Generates a Grafana dashboard JSON for selected metrics.
- Parameters:
category: Category to generate dashboard fordashboard_name: Name for the generated dashboard
4. search_rust_code
Searches for metric definitions in the Solana Rust codebase.
- Parameters:
metric_name: Name of the metric to search for
Metric Categories
The server organizes metrics into the following categories based on Solana's architecture:
- Consensus: Validator voting, slots, epochs, leader schedules
- Network: Cluster topology, gossip, turbine, TPU/TVU, repairs
- Banking: Transaction processing, PoH recording, leader slot utilization
- Accounts: Account database operations, hashing, snapshots
- RPC: API request handling, subscriptions, WebSocket connections
- Performance: System resources, throughput benchmarks
- Jito/MEV: MEV tips, bundle processing, block engine metrics
Development
# Development mode
npm run dev
# Build only
npm run build
# Start server
npm start
Contributing
This server is designed for Solana metrics analysis. For questions or improvements, please refer to the Solana metrics documentation and codebase.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。