$ loading_
Securely run local and remote commands via SSH with sessions and env variables.
This MCP tool claims local and SSH-based command execution, making it a sensitive system-operation tool; however, the materials show no built-in secret requirement, no explicit remote API endpoint, and it is open-source under MIT. Overall, it fits a caution-level profile, with the main concerns centered on command execution and the resulting local data access surface.
The materials state that no keys or environment variables are required, and there is no request for third-party API tokens. Note that its claimed SSH support may involve the user's own SSH credentials in practice, but this is not listed as a built-in requirement of the tool.
No fixed remote endpoint is listed by the system checks, but the description explicitly supports remote command execution over SSH, which implies commands and related input data may be sent to user-specified SSH hosts. There is no evidence that it exfiltrates data to unrelated third-party services.
The system has already flagged executes-code, and the description says it supports local and remote command execution with session management; this implies it can spawn local processes and invoke shell/system commands. This is a high-impact but standard capability for an MCP tool, and the current materials do not show abnormal privilege escalation or hidden execution beyond the stated function.
Any tool that can execute local commands can typically read/modify local files, environment variables, and other resources reachable under its runtime privileges; the stated environment variable support also increases potential exposure to sensitive configuration. The materials do not mention sandboxing or least-privilege controls, so it should be treated cautiously as having the same access as the local command context.
Positive factors include being open-source and MIT-licensed, making the code auditable in principle; however, it comes from a third-party registry, has 0 stars, unknown maintenance status, and no README, which limits confidence in maturity and verifiability. There are no clear red flags such as closed-source distribution, obvious deception, or malicious packaging, so it should not be elevated to risk.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "terminal-mcp-server" yet — see the docs or source repo.
Connect to the production server over SSH, check nginx and docker service status, and return the latest 50 lines of error logs.
Service status, issue summary, and relevant log output.
Set APP_ENV=staging on a remote Linux host, then sequentially pull code, install dependencies, restart the app, and keep the session output.
Step-by-step execution results, exit statuses, and full session logs.
Run a local system diagnostic script in the terminal to check disk, memory, and network connectivity, and output structured results.
Diagnostic metrics, a list of anomalies, and a readable summary of findings.
Securely let AI assistants run remote server commands over SSH.
Manage SSH sessions, run remote commands, transfer files, and diagnose servers.
Securely diagnose remote servers over SSH with guarded read and write access.
Connect AI to remote servers via SSH for commands and secure file transfer.
Let AI handle secure remote SSH commands, transfers, and port forwarding.
Give AI persistent SSH access for remote commands, file transfer, and port forwarding.