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 agentmods add instructions/aiondadotcom/mcp-ssh/claude-mdgit clone --depth 1 https://github.com/AiondaDotCom/mcp-sshWrote 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/instructions/aiondadotcom/mcp-ssh/claude-md)<a href="https://agentmods.dev/instructions/aiondadotcom/mcp-ssh/claude-md"><img src="https://agentmods.dev/badge/instructions/aiondadotcom/mcp-ssh/claude-md.svg" alt="Measured on agentmods" 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 | $0.03346 | $0.03346 |
| Opus 5 | $0.01673 | $0.01673 |
| Sonnet 5 | $0.00669 | $0.00669 |
| Haiku 4.5 | $0.00335 | $0.00335 |
Grade B, and why
mcp-ssh CLAUDE.md scanned grade B with 2 findings 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 4d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- The server enforces `chmod 600` on config files containing `@password` annotations Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- Node.js built-ins: `child_process`, `fs/promises`, `os`, `path` How it starts
The opening of the file, as written. The whole thing — 222 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
This is MCP SSH Agent (@aiondadotcom/mcp-ssh) - a Model Context Protocol (MCP) server that provides SSH operations for AI assistants like Claude Desktop. The project uses native SSH commands (ssh, scp) rather than JavaScript SSH libraries for maximum reliability and compatibility.
Development Commands
Basic Operations
npm start- Start the MCP server (same asnpm run dev)npm run dev- Start the MCP server with debug outputnpm run build- Compilesrc/todist/(tsc -p tsconfig.build.json)npm run typecheck-tsc --noEmitforsrc/(strict) and for the tests (relaxed)npm run lint/npm run lint:fix- ESLint with type-aware rulesnpm test- Run the vitest suite with coveragenpm run test:watch- Vitest in watch mode
The compiled output must exist before the server can run. npm install builds it via the
prepare script; after editing src/ run npm run build (or npm test, which reads src/
directly and needs no build).
Development Scripts
./start.sh- Start the server with debug output./start-silent.sh- Start the server in silent mode (no debug output)node bin/mcp-ssh.js- Direct server execution. It loadsdist/server.jsand fails with a clear message if the build is missing. Do not add anis this module run directly?check to the server module: that heuristic comparedprocess.argv[1]against forward-slash suffixes, never matched on Windows, and made the server exit silently (issue #8, regression in 1.3.8)
Publishing
npm version patch|minor|major- Bump version and create git tagnpm publish- Publish to npm (see PUBLISHING.md for details)npm pack- Create tarball for testing
MCP Bundle Building
npm run build:mcpb- Build the installable MCP Bundle (.mcpb) intobuild/./scripts/build-mcpb.sh- Direct build script execution
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.
- 4d ago First seen · 222 lines · 3,346 tokens per session scan B 2717af52ce2a
mcp-ssh CLAUDE.md is an instructions file published in the GitHub repository AiondaDotCom/mcp-ssh (96 stars, last pushed 24d ago), licensed MIT. It adds 3,346 tokens to every session, about $0.0167 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
mcp-ssh GEMINI.md
Gemini CLI instructions for zhouxiangjing/mcp-ssh, covering ssh mcp server — ai agent guide, role activation — read this first, critical: mcp trigger policy, tool groups & trigger conditions and 1. connection management.
Pake copilot-instructions.md
Copilot instructions for tw93/Pake, a project described as: 🤱🏻 Turn any webpage into a desktop app with one command.
swift-tui AGENTS.md
Instructions for SwiftTUI/swift-tui, covering agents.md, build & test commands, architecture (one-page summary), development guidelines and code style.
best-claude-hud AGENTS.md
Instructions for GaoSSR/best-claude-hud, covering project agent instructions and release work.
mcp-screenshot-server CLAUDE.md
Instructions for sethbang/mcp-screenshot-server, covering claude.md, what this is, commands, environment variables and architecture.
codebay CLAUDE.md
Instructions for khromov/codebay, covering claude.md, what this is, runtime & commands, comments and architecture.