Sshmcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Sshmcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A stateful SSH MCP server for Claude Code. Gives Claude persistent shell sessions over SSH — commands preserve working directory, environment variables, and state across calls.
& for long-running commandspip install -e .Add to ~/.claude/claude_desktop_config.json or your MCP settings:
{
"mcpServers": {
"ssh": {
"command": "sshmcp",
"args": []
}
}
}| Tool | Description |
|---|---|
ssh_connect | Connect to a host via SSH (password, key, or agent) |
ssh_exec | Run a command in the persistent shell |
ssh_read_output | Read recent terminal output |
ssh_upload | Upload a local file via SFTP |
ssh_download | Download a remote file via SFTP |
ssh_write_file | Write content to a remote file |
ssh_list_sessions | List all active sessions |
ssh_switch_session | Switch active session |
ssh_close | Close a session |
Once configured, Claude can:
ssh_connect(host="myserver.com", username="user", key_path="~/.ssh/id_ed25519")
ssh_exec(command="cd /opt/app && cat config.yaml")
ssh_exec(command="grep -r 'ERROR' /var/log/app/") # still in /opt/app context
ssh_upload(local_path="./exploit.py", remote_path="/tmp/exploit.py")~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.