ssh-mcp

ssh-mcp is a local MCP (Model Context Protocol) server that connects to a remote Linux/Windows host over SSH and exposes tools to run remote shell commands (exec and sudo-exec) from an MCP client.

Evaluated Mar 30, 2026 (22d ago)
Repo ↗ DevTools mcp ssh remote-exec automation typescript command-runner
⚙ Agent Friendliness
52
/ 100
Can an agent use this?
🔒 Security
50
/ 100
Is it safe for agents?
⚡ Reliability
38
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
75
Documentation
70
Error Messages
0
Auth Simplicity
55
Rate Limits
10

🔒 Security

TLS Enforcement
80
Auth Strength
55
Scope Granularity
15
Dep. Hygiene
55
Secret Handling
45

Communication with the SSH server is assumed encrypted (SSH), but TLS/transport specifics for the local MCP channel are not documented. The server supports password or SSH key auth to the target host. There is no documented MCP-level authorization, no scoped permissions, and no command allowlisting; this raises risk of arbitrary command execution if an agent is prompted with malicious instructions. Timeout and process abortion are mentioned for hanging commands, which is a helpful safety control. Secret handling details (e.g., whether args/logs expose passwords/sudo passwords) are not specified in the provided README.

⚡ Reliability

Uptime/SLA
0
Version Stability
55
Breaking Changes
50
Error Recovery
45
AF Security Reliability

Best When

You have an authenticated MCP client and want a convenient way to execute bounded, timeout-controlled SSH commands on a known set of servers.

Avoid When

When commands or connection parameters could be influenced by untrusted users, or when you need strict auditing, least-privilege command authorization, or tenant-level policy enforcement.

Use Cases

  • Letting an MCP-enabled agent run administrative or operational commands on remote servers
  • Remote execution via natural language using MCP tool calls
  • Automating recurring troubleshooting tasks over SSH
  • Controlled command execution with configurable timeout and input length limits

Not For

  • Unrestricted, internet-exposed remote command execution from untrusted inputs
  • High-assurance production automation without additional guardrails/auditing
  • Interactive long-running sessions that require streamed I/O beyond simple exec outputs
  • Use cases requiring robust tenant isolation or fine-grained per-command authorization

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: SSH password authentication SSH key authentication
OAuth: No Scopes: No

Authentication is to the remote host over SSH using a provided password or private key. There is no documented separate auth/authorization layer for the MCP client itself.

Pricing

Free tier: No
Requires CC: No

MIT-licensed OSS package; pricing for hosting/usage depends on your environment.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • sudo-exec requires sudoPassword and can be disabled entirely via --disableSudo
  • Command execution can be affected by --timeout (kills/aborts running processes) and --maxChars input length limits
  • No explicit allowlist/denylist or least-privilege model is documented—agents may be able to run arbitrary commands they are given

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for ssh-mcp.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered