System Metrics MCP Server

System Metrics MCP is a Python-based Model Context Protocol server that exposes system health, process information, disk details, network statistics, battery status, temperature sensors, and secure log file access.
- Install Python 3.11 or newer.
- Create and activate a virtual environment:

python -m venv .venv .\.venv\Scripts\activate

This starts the MCP server using stdio transport, which is compatible with Claude Desktop and other MCP clients.

Use the followingclaude.jsonor equivalent configuration block:

{ "mcpServers": { "system-metrics": { "command": "python", "args": ["server.py"] } } }

- get_system_status()
- list_high_cpu_processes(limit=5)
- list_high_memory_processes(limit=5)
- get_disk_details()
- read_log_file(filename)
- get_network_stats()
- get_battery_status()
- get_top_process_details(pid)
- get_system_temperature()

- read_log_fileonly allows files inside the local./logs/directory.
- Absolute paths and path traversal attempts are rejected.
- Log files larger than 1 MB are refused to prevent denial-of-service or memory exhaustion.
- MCP tool inputs are validated using FastMCP strict input validation.

- "What is the current CPU and RAM usage on the system?"
- "List the top 5 processes by CPU usage."
- "Show the top 5 memory-hungry processes."
- "Return the current disk utilization details."
- "Read the contents ofapp.logfrom the logs directory."
- "Provide the system network statistics."
- "What is the battery charge percentage and is the system plugged in?"
- "Give me detailed metrics for PID 1234."
- "Do we have temperature sensors available and what are their readings?"

- Log files can only be read from the local./logs/directory.
- Absolute file paths are rejected.
- Path traversal attempts such as../or../../are blocked usingpathlib.Path.resolve()and strict directory validation.
- Invalid file access attempts are logged and returned as human-readable errors.
- Each tool implements exception handling forFileNotFoundError,PermissionError,psutil.AccessDenied,psutil.NoSuchProcess, and unexpected exceptions.

Application logging is written tologs/server.logand also streamed to the console. The server logs:

- tool invocations
- access errors
- unexpected failures

- The server is designed for production-quality usage with async support and modular implementation.
- Thelogs/directory is intentionally included as an empty folder for secure log access and server logging.

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

An MCP server that gives AI agents like Claude, Cursor, and Gemini access to historical Windows CPU, GPU, temperature, and privacy data gathered by the AppControl Windows app.

Scout's official MCP pipes error, trace and metric data from production to your AI agent

Access and manage Grafana resources, including dashboards, datasources, Prometheus, Loki, and alerting.

A cross-platform server for real-time monitoring of CPU, GPU, memory, disk, network, and process information.

Expose Prometheus monitoring tools to an LLM for querying and analysis.

The IBM Instana MCP server enables seamless interaction with the IBM Instana observability platform, allowing you to access real-time observability data directly within your development workflow.

Real User Monitoring for Core Web Vitals. Query LCP, INP, CLS field data from real visitors at the p75 percentile.

Digi Remote Manager MCP allows users to connect Ai Agents to their Digi Remote Manager account for analyzing fleet data and help with troubleshooting.

Access Grafana resources like dashboards, datasources, Prometheus, Loki, and alerts.

MCP server for Grafana — manage dashboards, datasources, alerts, folders, and annotations over stdio