safe-ssh-mcp
Related Servers
Alternatives to safe-ssh-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityDmaintenanceA secure SSH-based MCP server for diagnosing remote servers. It allows AI agents to execute read-only commands and read files automatically, while requiring user confirmation for write operations.MIT
- AlicenseNot gradedqualityBmaintenanceMCP server for managing SSH connections to remote servers, providing host registry, session pool, and secure command execution with read-only protection.31 npmApache 2.0
- AlicenseNot gradedqualityDmaintenanceSSH-based MCP server that enables remote execution of SSH commands, file transfers, and secure server management via the MCP protocol.ISC
- AlicenseNot gradedqualityDmaintenanceA read-only MCP server for Linux and macOS system administration, diagnostics, and troubleshooting, supporting remote SSH execution and multi-host management.Apache 2.0
- AlicenseAqualityDmaintenanceAn MCP server for SSH/SCP operations with passwordless authentication, enabling remote command execution, file transfer, and session management.1916 npm1MIT
- AlicenseBqualityCmaintenanceA server that enables secure interaction with remote SSH hosts through standardized MCP interface, providing functions like listing hosts, executing commands, and transferring files using native SSH tools.7558 npm105MIT
TDQS
Scored across 55 tools
Most tools target a clearly distinct resource plus action, and descriptions actively disambiguate similar ones (e.g. get_disk_free vs get_disk_usage, get_pressure vs deprecated get_memory_pressure). A few pairs still require reading the description carefully (get_list_of_files vs get_list_of_files_with_filter, get_top vs get_ps_aux_top_cpu_consumers vs get_vmstat, the three disk tools), but no true duplicates.
Nearly everything follows a snake_case get_<resource> pattern (get_systemd_status, get_docker_inspect, get_crictl_ps, get_kubectl_logs), which is highly predictable. Minor deviations like check_tcp_connect and verbose names such as get_service_logs_from_journalctl or get_ps_aux_top_cpu_consumers are readable and still consistent in style.
55 tools is very heavy for a read-only SSH diagnostics server, with dense clusters of near-sibling commands (four crictl tools, four docker tools, four systemd list/status tools, three kubectl tools). Each is individually useful, but the surface is large enough that selection cost is high and some granularity could be consolidated.
Coverage of read-only Linux host diagnostics is broad and coherent: systemd, journald, docker, kubernetes/CRI, network reachability and routing, disk/inode/memory, package inventory, and login auditing. Gaps exist (no file-content read beyond lockfiles, no firewall/SELinux/certificate checks, no user/group inspection), but they are not fatal for the stated diagnostic purpose and the server is clearly read-only by design.