What can you do with this server?

The ssh-mcp-dynamic server enables secure, key-based SSH command execution on remote hosts, with support for both regular and sudo commands, configurable per-call parameters, and environment-variable-based defaults. * Execute shell commands: Run arbitrary shell commands on remote servers via SSH using the ssh_exec tool. * Execute privileged commands: Run commands with sudo privileges using the ssh_sudo_exec tool, which automatically prepends sudo. * Connect to multiple machines: Host, user, port, and key are all specified per call — not hardcoded — so a single server instance can target many machines. * Key-based authentication: Authenticate using PEM private keys via direct file paths (with ~ expansion) or pre-configured named shortcuts (e.g. prod, staging) defined in SSH_MCP_KEYS. No passwords are handled or stored. * Configurable defaults: Set default values for SSH key, username (default: root), port (default: 22), and timeout (default: 60000 ms) via environment variables to reduce repetition. * MCP client integration: Designed for use with MCP clients like Claude Desktop and Claude Code, allowing natural language prompts to invoke its tools.

How do I use ssh-mcp-dynamic?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@ssh-mcp-dynamic run 'uptime' on prod-web1" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

ssh-mcp-dynamic

by Calevi-Consulting

Overview Schema Related Servers Score Discussions

TypeScript

Local

ssh-mcp-dynamic

A minimal Model Context Protocol (MCP) server that lets an MCP client (e.g. Claude Desktop) run shell commands on remote hosts over SSH. The host, private key, user and port are chosen per call, so a single server instance can reach many machines.

It exposes two tools:

Tool	Description
`ssh_exec`	Run a shell command on a remote host.
`ssh_sudo_exec`	Run a shell command with `sudo` (don't include the `sudo` prefix yourself).

Authentication is key-based only (PEM private keys). No passwords are handled or stored.

Requirements

Node.js 18+
SSH access to the target hosts with a private key

Related MCP server: ssh-chat-mcp

Install & build

npm install
npm run build

This compiles src/index.ts to dist/index.js.

Configuration

Everything host-specific is supplied through environment variables — nothing is hardcoded in the source.

Variable	Default	Purpose
`SSH_MCP_KEYS`	`{}`	JSON object mapping key shortcuts to private-key paths. A leading `~` expands to the home directory.
`SSH_MCP_DEFAULT_KEY`	(none)	Shortcut or path used when a call omits `key`. If unset, `key` is required per call.
`SSH_MCP_DEFAULT_USER`	`root`	Default SSH username.
`SSH_MCP_DEFAULT_PORT`	`22`	Default SSH port.
`SSH_MCP_TIMEOUT_MS`	`60000`	Default command/connection timeout in milliseconds.

Example SSH_MCP_KEYS:

{
  "prod": "~/keys/prod.pem",
  "staging": "~/keys/staging.pem"
}

With that set, a call can pass "key": "prod" instead of a full path. You can also pass a full path directly at call time without configuring any shortcut.

Tool parameters

Both tools accept:

host (required) — IP or hostname.
command (required) — the shell command.
key — a configured shortcut or a path to the PEM file. Required unless SSH_MCP_DEFAULT_KEY is set.
user — SSH username (defaults to SSH_MCP_DEFAULT_USER).
port — SSH port (defaults to SSH_MCP_DEFAULT_PORT).
timeout — timeout in ms (defaults to SSH_MCP_TIMEOUT_MS).

Use with Claude Code (CLI)

The quickest way — no clone, no manual build. Claude Code runs it on demand via npx straight from GitHub.

Minimal — no environment config at all. You provide the host, command and a full key path on every call:

claude mcp add ssh-mcp -- npx -y github:Calevi-Consulting/ssh-mcp-dynamic

With shortcuts and defaults — preconfigure your keys once so calls can use a short name (e.g. prod) and omit the user/port:

claude mcp add ssh-mcp -s user \
  -e SSH_MCP_KEYS='{"prod":"~/keys/prod.pem"}' \
  -e SSH_MCP_DEFAULT_KEY=prod \
  -e SSH_MCP_DEFAULT_USER=ubuntu \
  -- npx -y github:Calevi-Consulting/ssh-mcp-dynamic

npx clones the repo, builds it (via the prepare script) and launches the server. Once published to npm you can drop the github: prefix and use npx -y ssh-mcp-dynamic.

Prefer a local checkout? Build it once and point Claude Code at the compiled file:

git clone https://github.com/Calevi-Consulting/ssh-mcp-dynamic.git
cd ssh-mcp-dynamic && npm install && npm run build

claude mcp add ssh-mcp -s user \
  -e SSH_MCP_KEYS='{"prod":"~/keys/prod.pem"}' \
  -e SSH_MCP_DEFAULT_KEY=prod \
  -- node "$(pwd)/dist/index.js"

Scopes (-s): local (default, current project only), user (all your projects), project (saved to a versioned .mcp.json to share with your team).

Verify with claude mcp list, or /mcp inside a session. Remove with claude mcp remove ssh-mcp.

Use with Claude Desktop

Add the server to your claude_desktop_config.json:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "ssh-mcp": {
      "command": "node",
      "args": ["/absolute/path/to/ssh-mcp-dynamic/dist/index.js"],
      "env": {
        "SSH_MCP_KEYS": "{\"prod\":\"~/keys/prod.pem\",\"staging\":\"~/keys/staging.pem\"}",
        "SSH_MCP_DEFAULT_KEY": "prod",
        "SSH_MCP_DEFAULT_USER": "ubuntu"
      }
    }
  }
}

Restart Claude Desktop after editing the config.

Usage

Once the server is registered, you don't call the tools directly — you ask your MCP client (Claude Code / Claude Desktop) in plain language and it invokes ssh_exec / ssh_sudo_exec for you. Some example prompts:

Using ssh-mcp, run `hostname && uptime` on 10.0.0.5 with the prod key.

Check the free disk space on staging.example.com (df -h) via ssh-mcp.

On 10.0.0.5, tail the last 50 lines of /var/log/syslog with sudo.

Restart nginx on web-01.example.com with sudo, then show `systemctl status nginx`.

Run `docker ps` on 203.0.113.10 as user ubuntu on port 2222 using ~/keys/prod.pem.

How those map to a tool call (the client fills this in for you):

// "run hostname on 10.0.0.5 with the prod key"
{
  "tool": "ssh_exec",
  "host": "10.0.0.5",
  "command": "hostname",
  "key": "prod"          // a configured shortcut, or a full path like ~/keys/prod.pem
}

// "tail syslog with sudo on 10.0.0.5"
{
  "tool": "ssh_sudo_exec",
  "host": "10.0.0.5",
  "command": "tail -n 50 /var/log/syslog"   // no 'sudo' prefix — the tool adds it
}

Tips:

Mention the host, the command, and which key/user/port when they aren't the configured defaults.
Naming the server ("using ssh-mcp…") helps the client pick the right tool when you have several MCP servers registered.
For privileged commands ask for "with sudo" so the client uses ssh_sudo_exec — and don't put sudo in the command yourself.

Security notes

This server executes arbitrary shell commands on remote hosts, including with sudo via ssh_sudo_exec. Only connect it to hosts and keys you control, and only run it with an MCP client you trust.
Private keys are read from disk at call time. Never commit private keys — *.pem, *.key, and common key filenames are already in .gitignore.
Prefer keys that are passphrase-protected or scoped to specific hosts.
The server communicates over stdio with the local MCP client; it does not open any network listener of its own.

License

MIT

Available Tools

2 tools

ssh_execC

Execute a shell command on a remote host via SSH.

ParametersJSON Schema

Name	Required	Description	Default
`key`	No	Provide a configured key shortcut or a full path to the private key file (supports a leading ~ for the home directory).
`host`	Yes	IP address or hostname of the remote server
`port`	No	SSH port (default: 22)
`user`	No	SSH username (default: root)	root
`command`	Yes	Shell command to execute on the remote server
`timeout`	No	Timeout in milliseconds (default: 60000)

TDQS

C2.9/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, and the description does not disclose behavioral traits such as potential side effects, security implications, or idempotency. For a command execution tool, this is a significant gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One concise sentence with no fluff. However, it could be more informative while remaining concise.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 6 parameters, no output schema, and no annotations, the description is too minimal. It fails to explain return behavior, error handling, or prerequisites, making it incomplete for safe and effective use.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema itself defines all parameters. The description adds no additional meaning or usage hints beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the action (execute) and resource (shell command on remote host via SSH). However, it does not explicitly differentiate from the sibling tool ssh_sudo_exec; the distinction is implied.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance on when to use this tool versus alternatives (e.g., ssh_sudo_exec). Lacks when-not or context for selection.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

ssh_sudo_execB

Execute a shell command with sudo on a remote host via SSH.

ParametersJSON Schema

Name	Required	Description	Default
`key`	No	Provide a configured key shortcut or a full path to the private key file (supports a leading ~ for the home directory).
`host`	Yes	IP address or hostname of the remote server
`port`	No	SSH port (default: 22)
`user`	No	SSH username (default: root)	root
`command`	Yes	Shell command to execute with sudo (do not include 'sudo' prefix)
`timeout`	No	Timeout in milliseconds (default: 60000)

TDQS

B3.3/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It discloses that the command runs with sudo but does not mention authentication requirements, key handling specifics, potential side effects, or error conditions. The disclosure is minimal beyond the core purpose.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, efficient sentence that conveys the core purpose without waste. It could include slightly more context (e.g., mentioning the key parameter or that 'sudo' should not be included in the command) without becoming verbose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 6 parameters, no output schema, and no annotations, the description is too brief to be complete. It lacks details about return values, error handling, authentication process, and usage caveats. The description only covers the basic purpose, leaving significant gaps for an agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema has 100% coverage with descriptions for all 6 parameters. The description adds no additional meaning beyond what the schema already provides. Baseline score is appropriate given the schema's thoroughness.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb 'Execute' and the resource 'a shell command with sudo on a remote host via SSH'. It distinguishes this tool from the sibling tool ssh_exec by explicitly mentioning 'with sudo', making its unique purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage when sudo is needed but does not provide explicit guidance on when to use this versus ssh_exec, nor does it mention prerequisites or scenarios where this tool should be avoided. The context is implied but not clearly stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

2 tool updatesv1.0.0Jul 14, 2026
- First observedssh_exec
- First observedssh_sudo_exec

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

B3.4/5.0

Disambiguation5/5

The two tools have clearly distinct purposes: one for regular commands, one for privileged commands with sudo. No ambiguity.

Naming Consistency5/5

Both tools follow a consistent 'ssh_verb' pattern with underscores, using 'ssh_exec' and 'ssh_sudo_exec'.

Tool Count4/5

Two tools is slightly below the typical range but appropriate for the narrow domain of remote command execution, covering the essential distinction.

Completeness4/5

The set covers the core operations for remote command execution (regular and sudo), but lacks additional utilities like file transfer or key management.

Maintenance

ActivityMaintained

ResponsivenessNo issues

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

termalin-web A
in.termal
Run commands and read/write files on your servers over Termalin's keyless tunnels (hosted MCP).
Shell Access File Systems
81
Turrior MCPOAuth
com.turrior.my
Authenticated remote MCP server for Turrior.
Cloud Platforms
3
supportsheep
io.github.pooriaarab
Remote MCP server for supportsheep: run AI interviews and manage support content for your blog.
Customer Support Content Management Systems Marketing
1
attest-mcp-remote A
io.github.SPAZIO-GENESI
Zero-install remote MCP server for proof-of-existence file attestation.
Blockchain Web3 & Decentralized Tech Security
82

Related MCP Servers

SSH MCP Server
Shell Access Command Line
mfangtao
A
license
B
quality
D
maintenance
A server that enables remote command execution over SSH through the Model Context Protocol (MCP), supporting both password and private key authentication.
Updated a year ago (2025-06-26 06:12 UTC)
1
1
2
MIT
ssh-chat-mcp
Shell Access Command Line
aiplatforms-ru
A
license
A
quality
D
maintenance
Zero-config SSH/SFTP MCP server that lets an LLM client open temporary SSH/SFTP sessions to remote hosts, run commands, and upload/download files without holding any pre-baked credentials.
Updated 4 months ago (2026-05-22 00:53 UTC)
17
10
2
MIT
ssh-client-mcp-server
Shell Access Command Line
jordanburke
A
license
Not graded
quality
A
maintenance
A local MCP server that enables LLMs to execute shell commands on remote hosts over SSH with multiple authentication methods.
Updated 2 months ago (2026-07-04 19:24 UTC)
35
3
MIT
mcp-ssh-wrapper
Shell Access Command Line
megascope
A
license
Not graded
quality
D
maintenance
A minimal MCP server that executes commands on remote hosts by delegating to the local ssh binary, supporting batch-mode execution and optional timeout.
Updated 6 months ago (2026-03-15 18:26 UTC)
MIT

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Calevi-Consulting/ssh-mcp-dynamic'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

ssh-mcp-dynamic

Requirements

Install & build

Configuration

Tool parameters

Use with Claude Code (CLI)

Use with Claude Desktop

Usage

Security notes

License

Available Tools

TDQS

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

SSH MCP Server

ssh-chat-mcp

ssh-client-mcp-server

mcp-ssh-wrapper

Latest Blog Posts

MCP directory API