"Shell" matching MCP servers:

Enables read-only observability of a Linux host via MCP, exposing allowlisted systemd, docker, nginx, logs, disk, and cert info without shell access.

Read-only MCP server to inspect allowlisted Docker containers, systemd services, JSONL logs, and HTTP health endpoints without arbitrary shell access.

Enables LLMs to monitor system health, manage processes, execute shell commands, and inspect environment variables and directories.

Enables AI agents to monitor server health and capacity, diagnose outages, inspect Docker and deployment status, and perform safe, bounded recovery actions via MCP without granting unrestricted shell or SSH access.

MCP server — read-only log access for AI assistants. Debug your Linux server with AI, without giving the AI shell access.

Provides read-only access to systemd units, failed units, timers, unit status and files, journal entries, boots, and system summaries through eight queries. Enables inspection and triage of systemd services without granting write access or shell execution.

Enables secure, read-only inspection of a VPS over SSH through approved operations such as system health, disk usage, container logs, and service status, without giving the AI unrestricted shell access.

Enables recording a Wine or macOS program on an Apple silicon Mac and analyzing its GPU usage — GPU milliseconds per frame, frame pacing, FPS and memory — through Instruments' Metal System Trace, and also reads Metal .gputrace captures without Xcode and reports Apple's published GPU family facts. Recording, listing, comparing, and paging through traces and captures are exposed as tools that a shell or an AI coding agent can call directly.

Enables agents to score, inspect, and gate tool calls before they execute, resolving shell commands the way a shell would in order to block destructive actions and detect multi-step attack sequences. It also redacts credentials and PII in either direction, requires signed single-use approval tokens for high-impact actions, and keeps a tamper-evident audit trail.

A read-only MCP server that lets an AI assistant inspect a single AWS lab instance's EC2 health, recent system/nginx errors from CloudWatch, and nginx service status via SSM, with strict allowlisting and no generic shell access.

Enables local LLM clients to capture and analyze network traffic using Wireshark's CLI tools, providing packet summaries, protocol hierarchies, conversations, endpoints, expert info, and stream reassembly without shell access.

Policy gateway for AI agent tool calls: checks every MCP, shell and file call against one provio.yaml (allow / deny / ask / redact) before it runs, pins MCP tool definitions against rug pulls, and records each decision in a tamper-evident, signed ledger.

Gives AI coding agents read-only access to Grafana's Loki logs, Prometheus metrics, Tempo traces and active alerts from within a project, using per-project context declared in the repo and one-command switching between test and prod environments. Credentials are discovered by parsing shell rc files directly, so it works even when spawned as a non-interactive MCP client process.

Enables MCP clients and AI agents to discover and invoke only explicitly authorized, resource-bounded operational capabilities through a policy-enforced gateway with confirmation, audit, and isolated executor routing. It separates trusted transport identity from execution and supports declared tools over stdio MCP without exposing generic shell or privileged API surfaces.

Enables AI agents to inspect and control a robotic VLA system via MCP, offering read-only tools for ROS topics, processes, tmux, and logs, as well as write tools for publishing prompts, issuing tmux commands, restarting services, and running shell commands inside containers.

A read-only MCP server for observing registered local and SSH hosts, providing system, service, container, Git, and GPU status/log inspection through a closed-world registry without arbitrary shell access.

A read-only, OAuth 2.1-authenticated MCP server that lets Claude inspect the live state of a self-hosted server — system resources, containers, service and website health, n8n automations, video jobs, backups, repositories, error summaries, a component design system, uptime monitoring, and its own audit trail. All twenty-one tools are strictly observation-only, with no shell, file browser, SQL console, or Docker socket, so nothing can be started, stopped, deployed, or deleted.