Ai Ssh Toolkit MCP Server
ai-ssh-toolkit
npm
v0.3.0
Source verified
Published by ebmarquez — publish provenance cryptographically ties this package to that repository. That is proof of origin, not an official vendor package.
MCP server for AI-driven SSH sessions with pluggable credential backends
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 1.2 = 99. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −1.2 | Reference to a credential-exfiltration path in metadata MTC-INJ-TARGET-1 | injection |
2. Client adoption risk — 99 − 6 = 93. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 11
high Shell/command execution in server code (dist/credentials/aws-secretsmanager.js)MTC-SRC-002
In the server's implementation (`dist/credentials/aws-secretsmanager.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { resolveCliPath } from "../utils/cli-resolver.js"; /** Promisified execF
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/aws-secretsmanager.js
high Shell/command execution in server code (dist/credentials/azure-keyvault.js)MTC-SRC-002
In the server's implementation (`dist/credentials/azure-keyvault.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { resolveCliPath } from "../utils/cli-resolver.js"; /** Promisified execF
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/azure-keyvault.js
high Shell/command execution in server code (dist/credentials/bitwarden.js)MTC-SRC-002
In the server's implementation (`dist/credentials/bitwarden.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { resolveCliPath } from "../utils/cli-resolver.js"; /** Promisified execF
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/bitwarden.js
high Shell/command execution in server code (dist/credentials/google-secret-manager.js)MTC-SRC-002
In the server's implementation (`dist/credentials/google-secret-manager.js:19`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from 'child_process'; import { promisify } from 'util'; import { resolve } from 'path'; const execFile
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/google-secret-manager.js
high Shell/command execution in server code (dist/credentials/hashicorp-vault.js)MTC-SRC-002
In the server's implementation (`dist/credentials/hashicorp-vault.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { resolveCliPath } from "../utils/cli-resolver.js"; /** Promisified execF
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/hashicorp-vault.js
high Shell/command execution in server code (dist/credentials/macos-keychain.js)MTC-SRC-002
In the server's implementation (`dist/credentials/macos-keychain.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { currentPlatform } from "../utils/platform.js"; import { resolveCliPath
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/macos-keychain.js
high Shell/command execution in server code (dist/credentials/onepassword.js)MTC-SRC-002
In the server's implementation (`dist/credentials/onepassword.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { resolveCliPath } from "../utils/cli-resolver.js"; /** Promisified execF
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/onepassword.js
high Shell/command execution in server code (dist/credentials/ssh-agent.js)MTC-SRC-002
In the server's implementation (`dist/credentials/ssh-agent.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { promisify } from "util"; const execFileAsync = promisify(execFile); /**
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/ssh-agent.js
high Shell/command execution in server code (dist/credentials/windows-credential.js)MTC-SRC-002
In the server's implementation (`dist/credentials/windows-credential.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execFile } from "child_process"; import { currentPlatform } from "../utils/platform.js"; /** Promisified execFile
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/credentials/windows-credential.js
high Shell/command execution in server code (dist/security/host-key-scanner.js)MTC-SRC-002
In the server's implementation (`dist/security/host-key-scanner.js:5`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: execFile, spawn } from 'child_process'; import { promisify } from 'util'; import { resolveCliPath } from '../utils/cli-r
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/security/host-key-scanner.js
low Reference to a credential-exfiltration path in metadataMTC-INJ-TARGET-1
Reference to a credential-exfiltration path in metadata detected in the description of tool "ssh_list_hosts". Instruction-like content in tool metadata is executed by the model, not the human, and is the primary tool-poisoning vector.
Evidence: ~/.ssh
Fix: Tool descriptions should describe behavior, not instruct the assistant. Treat imperative / secrecy / sequencing language in metadata as hostile.
Location: tool ssh_list_hosts · description
Tools 28
Each tool and what it can reach — statically extracted from the published source.
ssh_downloadingests untrusted inputcredential_diagnoseno sensitive capabilitycredential_getno sensitive capabilitycredential_list_backendsno sensitive capabilitycredential_map_reloadno sensitive capabilityssh_audit_log_readno sensitive capabilityssh_check_hostno sensitive capabilityssh_executeno sensitive capabilityssh_forward_closeno sensitive capabilityssh_forward_dynamicno sensitive capability
Show 18 more tools ↓
ssh_forward_listno sensitive capabilityssh_forward_localno sensitive capabilityssh_forward_remoteno sensitive capabilityssh_host_infono sensitive capabilityssh_host_key_listno sensitive capabilityssh_host_key_removeno sensitive capabilityssh_host_key_trustno sensitive capabilityssh_list_hostsno sensitive capabilityssh_multi_executeno sensitive capabilityssh_session_closeno sensitive capabilityssh_session_executeno sensitive capabilityssh_session_openno sensitive capabilityssh_sftp_listno sensitive capabilityssh_stream_cancelno sensitive capabilityssh_stream_listno sensitive capabilityssh_stream_readno sensitive capabilityssh_uploadno sensitive capabilityversion_checkno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (28 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 1
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v0.3.0 latest |
A 93/100 | 11 | 1.13.0 | 2026-09-07 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan ai-ssh-toolkit --onlineMore in Cloud & DevOps
Operator control CLI for AdGuard Home with an MCP adapter
interacting with Aliyun (Alibaba Cloud) services through This MCP server provides tools.
MCP server for the Aspro.Cloud REST API
MCP server for Atlassian Confluence and Jira Cloud APIs
AuroraCloud-first MCP server that exposes AuroraDocs workspace data to Claude Desktop and other AI agents
The official AWS Knowledge Base retrieval server using Bedrock Agent Runtime.