Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add KinoThe-Kafkaesque/ssh-mcp-server --skill ssh-remote-opsgit clone --depth 1 https://github.com/KinoThe-Kafkaesque/ssh-mcp-serverWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kinothe-kafkaesque/ssh-mcp-server/ssh-remote-ops)<a href="https://agentmods.dev/skills/kinothe-kafkaesque/ssh-mcp-server/ssh-remote-ops"><img src="https://agentmods.dev/badge/skills/kinothe-kafkaesque/ssh-mcp-server/ssh-remote-ops/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kinothe-kafkaesque/ssh-mcp-server/ssh-remote-ops"><img src="https://agentmods.dev/badge/skills/kinothe-kafkaesque/ssh-mcp-server/ssh-remote-ops.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00077 | $0.01885 |
| Opus 5.5 | $0.00031 | $0.00754 |
| Sonnet 5.5 | $0.00015 | $0.00377 |
| Haiku 4.5 | $0.00008 | $0.00188 |
Grade C, and why
ssh-remote-ops scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
add_credential { "name": "prod-web", "host": "203.0.113.10", "username": "deploy", "privateKeyPath": "~/.ssh/id_ed25519" } How it starts
The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SSH Remote Ops
Drive remote servers through the ssh-mcp-server MCP tools. Every remote action references a stored credential by name, so hosts and key paths never need to appear in the conversation after setup.
If the tools (ssh_exec, list_credentials, ...) are not available in this
agent, the MCP server is not installed. Walk the user through
references/setup.md and stop until it is connected.
Golden rules
- Start with
list_credentials. Pick the credential the user meant. If it is ambiguous (several hosts could match "the server"), ask - do not guess. - Read before you write. Inspect state (status, disk, logs, config) before changing anything, and show the user what you found.
- Confirm before changing a server. Restarting or stopping services, editing config, deleting or overwriting files, package upgrades, firewall changes and anything on a production host need an explicit "yes" from the user first. Say exactly which command will run on which credential.
- Prefer
ssh_exec_rawwhen the command has quotes,|inside a regex,$, globs you want passed literally, or user-supplied values. Each array element reaches the remote program as one argument - no escaping bugs, no injection. - Bound your output. Use
tail -n,head,--since,-nandgrep -mso results stay readable. Output over 100 KB is cut to its first and last 50 KB. - Never paste secrets into commands. Don't echo private keys, tokens or
.envcontents back to the user unless they asked for that specific value.
Choosing the tool
| Need | Tool |
|---|---|
| One-off command, pipes and shell features welcome | ssh_exec (runs via bash -ic, so the user's aliases and PATH apply) |
| Command with tricky quoting or untrusted values | ssh_exec_raw with an argv array |
| Long command (build, backup, migration) | either exec tool with timeout up to 600000 ms |
| Copy one file | scp_copy (SFTP, no local binaries needed) |
| Copy a directory or many/large files | rsync_copy (incremental; needs rsync + ssh locally) |
Program that prompts, or a REPL / TUI (htop, psql, mysql, password prompt) |
ssh_session_* |
| Reach a remote-only port (DB, admin UI) from this machine | ssh_tunnel_start type local |
| Let the server reach a port on this machine | ssh_tunnel_start type remote |
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 141 lines · 77 tokens per session scan C fc2e699caf4e
ssh-remote-ops is a skill published in the GitHub repository KinoThe-Kafkaesque/ssh-mcp-server (11 stars, last pushed 10d ago), licensed MIT. It adds 77 tokens to every session and 1,885 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-28.
ssh-remote-ops compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ssh-remote-opsKinoThe-Kafkaesque/ssh-mcp-server | Skill | 11 | 10d ago | MIT | C | 77 | Claude Code |
| agentcore-investigationawslabs/mcp | Skill | 9.9k | yesterday | Apache-2.0 | A | 52 | Claude Code, Codex |
| frontmcp-configagentfront/frontmcp | Skill | 148 | yesterday | Apache-2.0 | A | 187 | Claude Code |
| ssh-incidentbvisible/mcp-ssh-manager | Skill | 533 | yesterday | MIT | A | 46 | Claude Code, Codex |
| frontmcp-setupagentfront/frontmcp | Skill | 148 | yesterday | Apache-2.0 | A | 176 | Claude Code, Codex |
| cis-aws-foundations-6.2CyberStrikeus/CyberStrike | Skill | 3.4k | 4d ago | AGPL-3.0 | A | 33 | Claude Code, Codex |
| cis-aws-foundations-6.4CyberStrikeus/CyberStrike | Skill | 3.4k | 4d ago | AGPL-3.0 | A | 25 | Claude Code, Codex |