@yawlabs/ssh-mcp

One click adds this to your local Yaw MCP config so it's available in every Yaw Terminal session. Or install manually below.

Make SSH work for AI tools. MCP server that manages your SSH environment, diagnoses what's broken, fixes it, and gives your agent remote access to anything.

Built and maintained by Yaw Labs.

The problem

AI CLI tools run in subprocesses where SSH is constantly broken. The agent tries to git pull and gets Permission denied (publickey). It tries to SSH into a server and the agent socket is stale. It tries to deploy and the host key changed because the instance was recreated. Every time, the AI has no idea what's wrong and spirals.

This happens across every situation that needs SSH keys:

• Git — clone, pull, push, fetch, submodules, LFS
• Package managers — npm install, pip install, go get, cargo, composer from private repos
• Server access — SSH, SCP, SFTP, rsync
• Tunneling — port forwarding to databases, SOCKS proxies
• Deployment — Ansible, Terraform, Capistrano, deploy scripts
• Cloud — AWS EC2, GCP, Azure, DigitalOcean, any VPS

ssh-mcp fixes this. It manages the SSH agent, loads keys, diagnoses failures with actionable fix commands, and provides remote operations — all as MCP tools your AI agent can call.

Quick start

Add to your MCP client config:

{
  "mcpServers": {
    "ssh": {
      "command": "npx",
      "args": ["-y", "@yawlabs/ssh-mcp@latest"]
    }
  }
}

On Windows wrap with cmd /c since Node 20+ can't spawn .cmd files directly:

{
  "mcpServers": {
    "ssh": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "@yawlabs/ssh-mcp@latest"]
    }
  }
}

The @latest tag makes npx re-resolve against the registry on every spawn, so each MCP session uses the newest published version. Or install globally if you'd rather pin (no auto-update):

npm install -g @yawlabs/ssh-mcp
# then in client config: "command": "ssh-mcp"

Tools

SSH environment management

Tools that fix your local SSH setup so everything else — git, deploys, tunnels — stops breaking.

Diagnostics

Remote operations

Higher-level operations

Tools that wrap common patterns agents build with ssh_exec — faster and less error-prone.

Auto-diagnostics

When any remote operation fails, ssh-mcp automatically runs diagnostics and includes the results in the error response. Your agent doesn't need to call ssh_diagnose separately — it gets told what's wrong and how to fix it right in the error message.

Connection pooling

Remote operations reuse SSH connections automatically. When your agent makes multiple calls to the same host, the first call opens a connection and subsequent calls reuse it. Connections are kept alive for 60 seconds after the last use, then closed automatically.

The pool caps at 100 connections by default, and a connection still being dialed counts against the cap. When the cap is reached the pool closes an idle connection to make room. If every slot is in use or dialing, a tool call waits for one instead of failing:

• A single-host remote tool (ssh_exec, the SFTP tools, ssh_find, ssh_tail, ssh_service_status) waits up to its timeout (30s for the SFTP tools, which have no timeout parameter) and is next in line for the first slot that frees, so an ssh_exec arriving in the middle of a wide ssh_multi_exec runs between two of its hosts rather than being refused for the whole run. It reports Connection pool is full only if it never won a slot in that time.
• ssh_multi_exec runs at most SSH_MCP_MAX_POOL_SIZE hosts at once (default 100) and works through a longer list as slots free up, so raising the cap raises its parallelism; timeout is per host, not a bound on the whole call. When the pool is full it waits, and gives up only when none of its own hosts holds a slot and a full timeout has passed with none of them starting or finishing. The hosts waiting at that point (up to one per parallel slot) and every host still queued then report Connection pool is full; the queued ones are never attempted.

A call reaches Connection pool is full only when other calls — a long ssh_exec, a concurrent ssh_multi_exec — kept every slot from it for a full timeout; a fan-out wider than the cap that has the pool to itself never does, however long it runs. If you see it, rerun once those calls finish (for ssh_multi_exec, the hosts that reported it), or raise the cap with SSH_MCP_MAX_POOL_SIZE=<n> in the server's environment. The error names the variable.

The library's acquire() / withConnection() do not wait by default: they reject at once with a PoolFullError (code: "ERR_SSH_MCP_POOL_FULL", isPoolFullError()) unless you pass { waitForCapacityMs } — see Programmatic usage.

SSH config support

All connections respect your ~/.ssh/config. Host aliases, custom ports, usernames, identity files, and ProxyJump settings are used automatically. If you have Host myserver configured in your SSH config, just pass host: "myserver" — ssh-mcp resolves everything.

ProxyJump / bastion hosts are supported automatically. If your SSH config has ProxyJump bastion for a host, ssh-mcp connects through the bastion transparently. Chained proxies work too.

Host key verification

All remote operations verify the server's host key against ~/.ssh/known_hosts:

• Known host, key matches — accept.
• Known host, key changed — reject (MITM protection). The rejection message distinguishes a genuine key mismatch from "the server offered a key type your known_hosts entry doesn't cover", so a missing ed25519 line doesn't read as an attack.
• Unknown host — accept, unless SSH_MCP_STRICT_HOST_KEY=1.

That last branch is trust-always, not TOFU. Real trust-on-first-use pins the key it saw the first time and rejects a change afterwards. The connection path never writes to known_hosts — no tool adds an entry as a side effect of connecting — so connecting pins nothing: every connection to a host absent from known_hosts is a "first" use and is accepted, including one where an attacker swapped the key since your last call. Only hosts put into known_hosts out of band get mismatch protection — by you, by ssh-keyscan, or by ssh_known_hosts_fix, the one tool here that does write the file. Call it explicitly to add an entry so future changes are caught.

For stricter environments, set SSH_MCP_STRICT_HOST_KEY=1 to reject unknown hosts. Add them explicitly with ssh_known_hosts_fix first.

The diagnostic tools (ssh_test, ssh_diagnose) use StrictHostKeyChecking=no for their probe commands. Those probes only run echo SSH_OK — no credentials or data pass through — so the relaxed setting is safe for connectivity testing. Real operations always go through the hostVerifier.

Command policy

ssh_exec and ssh_multi_exec accept free-form shell commands from the agent. For security-conscious deployments, you can restrict which commands run via two env vars, each accepting a comma-separated list of regex patterns:

• SSH_MCP_COMMAND_WHITELIST — if set, the command must match at least one pattern, else it's blocked.
• SSH_MCP_COMMAND_BLACKLIST — if set, the command must not match any pattern, else it's blocked.

When both are set, the command must pass both checks (whitelist first, then blacklist). When neither is set (the default), all commands are allowed.

Patterns are JavaScript regexes. Use ^ and $ for anchored matches; otherwise patterns are treated as substring matches. Commas are the delimiter, so a literal comma in a pattern needs to be expressed as \x2c or via a character class.

# Read-only allowlist: only ls / df / cat / find / tail
SSH_MCP_COMMAND_WHITELIST="^ls( .*)?,^df( .*)?,^cat ,^find ,^tail "

# Block destructive ops even if your agent goes off-script
SSH_MCP_COMMAND_BLACKLIST="^rm ,^shutdown,^reboot,^mkfs,^dd if=,>\s*/dev/"

Blocked commands surface as a clear error mentioning which pattern (or which env var) rejected the call, so the agent can adapt rather than guess. Policy is enforced before the SSH connection opens — no remote process is started for a blocked command.

Scope: policy covers ssh_exec and ssh_multi_exec only

Every other tool runs unchecked. That splits into two very different cases.

The structured read tools are all exempt, but for two different reasons — they don't reach the remote the same way.

ssh_find, ssh_tail and ssh_service_status do build a shell command (find, tail, systemctl), but from typed parameters with every interpolated value shell-quoted, never from free-form agent input. They're exempt for ergonomics: a tight ^ls whitelist would otherwise force you to allow ^find , ^tail , ^systemctl just to keep those tools working — defeating the point of a tight whitelist.

ssh_ls, ssh_stat, ssh_read_file and ssh_download build no command at all. They're pure SFTP (readdir, stat, readFile, fastGet), so — exactly like the mutating SFTP tools below — there is no command string for a regex to match, and these env vars could not gate them even if you wanted them to. They're grouped with the reads rather than flagged as a gap because they don't mutate remote state. One caveat: ssh_download is non-mutating on the remote only — it writes to whatever local path it's handed, and these env vars don't constrain that either.

The SFTP tools that mutate remote state are also unchecked, and that is a genuine gap — not an ergonomics call. ssh_write_file, ssh_upload, ssh_mkdir, and ssh_delete never build a shell command string, so a command-shaped regex has nothing to match. Concretely: SSH_MCP_COMMAND_BLACKLIST="^rm " does not stop ssh_delete, and SSH_MCP_COMMAND_WHITELIST="^ls " does not stop ssh_write_file. Closing this would need a separate path-policy mechanism, which this server deliberately does not have. If you must prevent remote mutation, drop those four tools from your MCP client's tool allowlist (or run a client that gates them) — these two env vars cannot do it.

Policy interaction with the env parameter (ssh_exec, ssh_multi_exec)

When ssh_exec or ssh_multi_exec is called with env: { KEY: "value" }, the values are injected as a KEY='value' ... shell prefix before the command (see the tool descriptions). Policy is checked against the full prefixed command, not the bare command argument — once, before fan-out, in the ssh_multi_exec case. That's the safer ordering at the protocol layer — but it means whitelist patterns need to anticipate the prefix and must be anchored, not substring matches:

# WRONG -- blocks any ssh_exec call that uses `env`, because the final command
# starts with `KEY='value' ` and never matches `^ls`.
SSH_MCP_COMMAND_WHITELIST="^ls "

# RIGHT -- allow zero or more `KEY='value' ` prefixes before the real command.
SSH_MCP_COMMAND_WHITELIST="^([A-Za-z_][A-Za-z0-9_]*='[^']*' )*ls( |$)"

You don't have to diagnose this from a bare rejection: when a whitelist blocks a call that used env, the error appends a note explaining that the prefix is why the ^ anchor stopped matching, and suggests the tolerant pattern above.

Avoid substring-match patterns like ls if you're worried about a hostile agent. An agent could pass env: { ATTACK: " ls " } to make the final command ATTACK=' ls ' rm -rf /, which matches a substring ls and bypasses the whitelist. Anchored patterns of the form above don't have this weakness because they require the real command name to follow the env-prefix block, not appear inside a quoted env value.

Blacklists need the same care. ^rm blocks a bare rm call, but doesn't block FOO='bar' rm. Use the same env-prefix-tolerant anchor:

SSH_MCP_COMMAND_BLACKLIST="^([A-Za-z_][A-Za-z0-9_]*='[^']*' )*rm( |$)"

If you don't trust the agent's env values at all, the simplest mitigation is to leave env unused in your client config and pass everything through the command string yourself.

Windows support

On Windows, ssh-mcp uses the OpenSSH Authentication Agent's \\.\pipe\openssh-ssh-agent named pipe automatically when SSH_AUTH_SOCK is not set. No SSH_AUTH_SOCK needed — just make sure the OpenSSH agent service is running.

ssh_agent_ensure and ssh_diagnose probe that pipe and tell you if the service is down. Remote operations do not: they assume the pipe and let the connection fail on its own if the agent isn't there. That is why a stopped agent service shows up as an auth failure rather than an "agent not running" error until you run the diagnostic tools.

Authentication

All remote operations accept connection parameters:

Auth resolution. An explicit credential wins outright and nothing else is offered. With neither given, ssh-mcp offers the ssh-agent and one on-disk key together — the way the OpenSSH client does — and lets the server pick during the auth exchange. It is not a strict first-match chain past step 2.

1. Explicit privateKeyPath — used alone. The agent is not offered and no other key is read.
2. Explicit password — used alone, same as above.
3. Neither given — both of the following are configured on the same connection:
   • ssh-agent — SSH_AUTH_SOCK, or on Windows the \\.\pipe\openssh-ssh-agent named pipe. The Windows pipe is assumed unconditionally; ssh-mcp does not check that the OpenSSH Authentication Agent service is actually running.
   • One on-disk key — the first readable path in ssh -G <host>'s identityfile list. OpenSSH emits that list for every host, including one with no IdentityFile line (it defaults to ~/.ssh/id_rsa, id_ecdsa, id_ecdsa_sk, id_ed25519, id_ed25519_sk), so this is the normal path — not a path reserved for hosts you configured an identity for. ssh-mcp's own built-in list (~/.ssh/id_ed25519, id_rsa, id_ecdsa) is a fallback used only when ssh -G cannot run at all, e.g. no SSH client installed.

Two things worth knowing about step 3:

• Only one on-disk key is ever offered — the first candidate that exists. ssh-mcp does not walk the whole identity list the way ssh does, so if the first readable key is the wrong one, authentication rests on the agent's keys. Pass privateKeyPath to force a specific key.
• When an agent is configured, an encrypted on-disk key is skipped and the scan moves to the next candidate. The underlying ssh2 library parses privateKey eagerly and errors with "no passphrase given" on an encrypted key, which would break the common setup of an encrypted key on disk with its decrypted copy loaded in the agent. With no agent configured, the first existing key is loaded regardless of encryption and ssh2 surfaces the passphrase error itself. Because the Windows pipe above is assumed unconditionally, the skip is always in force on Windows.

Example workflows

Agent can't git pull

Agent calls ssh_git_check → "Permission denied. Your SSH key is not registered with github.com."
Agent calls ssh_key_list → finds id_ed25519 exists but is not loaded
Agent calls ssh_key_load("~/.ssh/id_ed25519") → "Key loaded"
Agent calls ssh_git_check → "Git SSH authentication to github.com succeeded as username"
Agent runs git pull → works

Host key changed after instance recreation

Agent calls ssh_exec on server → error: "Host key verification failed"
  (auto-diagnostics included in error: "Fix with ssh_known_hosts_fix")
Agent calls ssh_known_hosts_fix("my-server") → "Host key refreshed"
Agent calls ssh_exec → works

First-time connection to a new server

Agent calls ssh_test("new-server") → "Connection refused at new-server:22"
Agent calls ssh_diagnose("new-server") → full report showing agent running, keys loaded, but host unreachable
Agent reports: "SSH server isn't running on new-server or port 22 is blocked"

Programmatic usage

import { connect, exec, diagnose, ensureAgent, listSshKeys, checkGitSsh, ConnectionPool, isPoolFullError } from '@yawlabs/ssh-mcp';

// Fix SSH environment
const agent = ensureAgent();
console.log(agent.message);

// Check git access
const git = checkGitSsh('github.com');
console.log(git.message);

// List available keys
const keys = listSshKeys();
for (const key of keys) {
  console.log(`${key.name} (${key.type}) - ${key.loadedInAgent ? 'loaded' : 'not loaded'}`);
}

// Run a remote command (one-off)
const client = await connect({ host: 'my-server', username: 'deploy' });
const result = await exec(client, 'uptime');
console.log(result.stdout);
client.end();

// Run multiple commands with connection pooling
const pool = new ConnectionPool();
await pool.withConnection({ host: 'my-server' }, async (client) => {
  const r1 = await exec(client, 'uptime');
  console.log(r1.stdout);
});
// Connection stays open for 60s — next call reuses it
await pool.withConnection({ host: 'my-server' }, async (client) => {
  const r2 = await exec(client, 'df -h');
  console.log(r2.stdout);
});
// A full pool rejects at once with a PoolFullError; pass a budget to wait for a slot
// instead, the way the tools do, and match the rejection on its code, not its text
try {
  await pool.withConnection(
    { host: 'other-server' },
    async (client) => {
      const r3 = await exec(client, 'uptime');
      console.log(r3.stdout);
    },
    { waitForCapacityMs: 30_000 },
  );
} catch (err) {
  if (!isPoolFullError(err)) throw err;
  console.log('no slot within 30s:', err.message);
}
pool.drain(); // close all connections when done

// Diagnose issues
const report = diagnose('my-server');
console.log(report.overall); // "ok" | "warning" | "error"
for (const check of report.checks) {
  console.log(`[${check.status}] ${check.name}: ${check.message}`);
}

Runtime selection

The ssh-mcp command — what npx @yawlabs/ssh-mcp runs — is a small launcher. It prefers the oam runtime and otherwise runs the server on Node, so oam is optional. Its minimum is the latest oam release, currently 0.15.2, and it never serves on an older oam.

Two environment variables control the choice; set them in your MCP client's env block:

• SSH_MCP_RUNTIME — which runtime to use. Case-insensitive; any value other than oam or node behaves as auto.
  • auto (default) — use the newest oam it can find at 0.15.2 or newer, otherwise run on Node.
  • oam — require oam. Fails loudly: if no usable oam is found, the launcher prints what it found and why each one was passed over to stderr, and exits with status 1 instead of starting the server.
  • node — always Node. The launcher does not look for oam, and OAM_BIN is ignored.
• OAM_BIN — path to an oam binary to use in preference to discovery, when it is 0.15.2 or newer. If it does not exist, is older, or will not run, the launcher says so on stderr and carries on with discovery.

Discovery looks in %LOCALAPPDATA%\oam\bin (Windows only), then ~/.oam/bin, then every directory on PATH; asks every oam binary it finds for its version; and uses the newest one at 0.15.2 or newer. On a tie the one found first wins, so an installed copy beats one on PATH. On Windows only oam.exe counts: an oam.cmd/oam.bat shim on PATH is never run, and is named on stderr when no usable oam is found.

By default an unusable oam is not an error. When Node starts the launcher (npx or a global install, run directly by your MCP client), auto mode falls back to Node — silently when there was nothing to find, or with a note on stderr naming each oam (and OAM_BIN) it passed over and why: older than 0.15.2, not runnable, or a shim. A chosen oam that then fails to start is named on stderr too. When an older oam starts the launcher, the handoff to Node is always noted on stderr (see below). With SSH_MCP_RUNTIME=oam, each of these cases exits with status 1 instead.

When the launcher is itself started by oam (oam run <path>/bin/ssh-mcp.mjs, which is how Yaw MCP starts an npx @yawlabs/ssh-mcp entry when a recent oam is installed):

• On oam 0.15.2 or newer it runs the server on that oam, in the same process. Nothing is discovered or spawned, OAM_BIN is not read, and SSH_MCP_RUNTIME=oam counts it as the oam it requires.
• On an older oam it never runs the server there. It hands the server off to the newest usable oam it can find, or to Node on PATH, or exits with status 1 when there is neither (SSH_MCP_RUNTIME=oam exits with status 1 rather than handing off to Node).
• With SSH_MCP_RUNTIME=node it hands the server off to Node on PATH, whatever the oam version.

For example, to always run on Node:

{
  "mcpServers": {
    "ssh": {
      "command": "npx",
      "args": ["-y", "@yawlabs/ssh-mcp@latest"],
      "env": { "SSH_MCP_RUNTIME": "node" }
    }
  }
}

On Windows, add the same env block to the cmd /c form from Quick start.

Requirements

• Node.js 18+
• SSH client installed (for diagnostics and environment management)
• Optional: oam 0.15.2+ — see Runtime selection

License

MIT