ssh_exec · Self-hosted MCP server for Claude MCP

WHAT IT DOES

What ssh_exec does on Self-hosted MCP server for Claude

AI agents invoke ssh_exec to trigger actions in Self-hosted MCP server for Claude. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why ssh_exec needs a policy

This tool enables execution of arbitrary shell commands on remote servers via SSH. The blast radius is critical because: (1) shell commands can perform any action the SSH user account permits, including reading sensitive data, modifying system files, installing malware, or disrupting services; (2) the AI agent controls the command content; (3) effects are irreversible and depend on agent reasoning.

From the tool's definition Tool description explicitly states 'Run a shell command on a remote host over SSH' — this executes arbitrary shell commands on remote systems with effects determined entirely by the command arguments passed by the AI agent.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

ssh_exec Rate-limited

ssh_exec stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "ssh_exec": {
      "limits": [
        {
          "counter": "ssh_exec_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Self-hosted MCP server for Claude, so you can run agents against it safely.

FAQ

Questions about ssh_exec

What does the ssh_exec tool do? +

Run a shell command on a remote host over SSH. Two ways to specify the target:\n\n(A) PREFERRED: pass host=<name from hosts.json> only — no. It is categorised as a Execute tool in the Self-hosted MCP server for Claude MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on ssh_exec? +

Register the Self-hosted MCP server for Claude MCP server in PolicyLayer and add a rule for ssh_exec: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Self-hosted MCP server for Claude. Nothing to install.

What risk level is ssh_exec? +

ssh_exec is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit ssh_exec? +

Yes. Add a rate_limit block to the ssh_exec rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block ssh_exec completely? +

Set action: deny in the PolicyLayer policy for ssh_exec. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides ssh_exec? +

ssh_exec is provided by the Self-hosted MCP server for Claude MCP server (leszczynskikarol/mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.