Full SSH access for AI agents with secret-safe environment injection and automatic output redaction. Persistent sessions, structured execution, SFTP, port forwarding — secrets are fed via stdin (invisible to `ps`) and scrubbed from all tool responses before reaching the LLM. `uvx mcp-remote-ssh`
Example client config (verify package name in the repo README)
{
"mcpServers": {
"mcp-remote-ssh": {
"command": "uvx",
"args": ["mcp-remote-ssh"]
}
}
}Transparent heuristic — same formula for every entry. Total 53/100.
Manage multiple SSH servers from one MCP: command execution, file transfer/rsync, database dump/import/query (MySQL/PostgreSQL/MongoDB), backups & restore, health monitoring, persistent sessions, tunnels, ProxyJump/bastion, and opt-in per-server security modes (readonly/restricted) with audit log. Linux, macOS & Windows OpenSSH.
Manage AWS EC2, Hetzner, OVH and custom SSH servers: run commands, fetch logs, CloudWatch/CloudTrail queries, fleet health snapshots, IP banning, S3 — with readonly/standard/dangerous guard tiers and a JSONL audit trail.
Fast SSH connection manager with TUI dashboard and MCP server for discovering, searching, and executing commands on remote hosts.
SSH & Kubernetes access broker where the model never touches a credential: a separate signer mints per-operation ephemeral SSH certificates and bound ServiceAccount tokens, with a per-command allow/deny/approval firewall (POSIX-sh AST), dry-run, human-in-the-loop approvals, session recording and an Ed25519-chained audit log. Self-hosted Go binaries, GPLv3.
Local-first CLI and MCP wrapper for AI coding agents. Routes shell commands through `sage run --`, stores command history locally, and returns compressed terminal output to reduce context noise.
Security-hardened MCP server for Linux system administration via 189 typed actions instead of shell strings, with an Ed25519-signed hash-chain audit log, one-time TTL approval receipts, and automatic rollback. Works with Claude Code, Cursor, and Codex CLI.