Node Command

io.github.huseyincigPublished by GitHub account huseyincigv0.26.125Updated Oct 25, 2026

Remote MCP server exposing your Node Command fleet (files, shell, processes, browser) to AI hosts.

VerifiedStreamable HTTPWeb executableCloud & InfrastructureBrowser AutomationDeveloper Tools

Overview

AI-generated overview

Remote MCP server that exposes a Node Command fleet — files, shell, processes and browser — to AI hosts so an assistant can run commands and manage hosts.

What you can do
  • List and resolve target hosts in the fleet
  • Run shell commands and declared project tasks
  • Manage long-running processes and poll job status
  • Launch browser or desktop GUI sessions and drive automated web interaction
  • Run read-only system snapshots and diagnostics before acting
When to use it
Useful when an assistant should operate machines in a Node Command fleet: running commands, starting project tasks, managing processes, or driving a browser on a remote host, with read-only checks before riskier steps.
Requirements
A remote endpoint at mcp.nodecommand.app; no package install. The manifest declares no authentication, environment variables or headers, but the hosted platform filters tool discovery by role and scope, so a Node Command account and host access are implied.
Before you install
Gives an assistant shell execution, process control and GUI/browser control on your hosts; destructive or high-risk actions are expected to need explicit confirmation. Tool availability depends on the connected user's role and scope, so check the authenticated tool list rather than assuming every tool is callable.

Installation

In SourceWeft

  1. Open Node Command in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.huseyincig%2Fnc-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "node-command": {
      "type": "http",
      "url": "https://mcp.nodecommand.app/mcp"
    }
  }
}

README

nc-protocol

Public wire contracts, Model Context Protocol (MCP) tool schemas, and agent protocol definitions for Node Command.

This repository is the canonical source of truth for the public Node Command contracts. It contains strictly declarative schemas and contracts: no proprietary server code, no database schemas, and no secrets.

Consumers — the Node Command hub, the nc-agent runtime, and client integrations — vendor this repository at a pinned commit, so contract changes here are explicit and auditable.

Namespaces

text
agent/     Hub ↔ Agent wire protocol (hello / welcome / request / reply / heartbeat)mcp/       Canonical authoring definitions and contract manifestpolicy/    Public host-policy projection + capability namesbrowser/   Public browser and egress policy schemasbindings/  Generated language bindings (JavaScript ESM)

Consuming the contracts

There is no runtime dependency to install: consumers vendor this repository at a pinned commit SHA and import from the vendored tree.

bash
git clone https://github.com/huseyincig/nc-protocol.git

Import the JavaScript bindings directly:

javascript
import {  MCP_TOOL_CONTRACT_VERSION,  MCP_TOOL_MANIFEST_FINGERPRINT,  MCP_TOOL_NAMES,  HOST_POLICY_CAPABILITY_NAMES} from './nc-protocol/bindings/javascript/contracts.mjs';

JSON schemas can be imported by path:

javascript
import helloSchema from './nc-protocol/agent/hello.schema.json' with { type: 'json' };import toolsIndex from './nc-protocol/mcp/tools/index.json' with { type: 'json' };

The current contract version, fingerprint, and capability list are pinned in mcp/contract-version, mcp/fingerprint, and policy/capability-names; the current generation and rollout rules are in COMPATIBILITY.md.

The manifest counts canonical authoring definitions, not the number of tools available to a particular connected user. The hosted platform composes these definitions into its public nc_ surface and filters discovery by role and scope. Use that server's authenticated tools/list response for callable names and schemas; do not invoke an authoring identifier merely because it appears here. Internal operation identifiers and the agent wire protocol remain distinct.

Testing & Validation

nc-protocol has zero runtime dependencies and uses the native Node.js test runner:

bash
# Clone the repositorygit clone https://github.com/huseyincig/nc-protocol.gitcd nc-protocol
# Run test suitenpm test# Or:node --test tests/*.mjs

The test suite validates:

  1. Presence of all required namespace files.
  2. Exact matching between tool definitions, mcp/tools/index.json, and tool count.
  3. Cryptographic consistency of the canonical sha256 manifest fingerprint.
  4. Schema parse validity and schema draft conformance.
  5. Binding parity with declarative contract files.

Contributing

See CONTRIBUTING.md for detailed guidelines on how to propose new tools, update schemas, recompute fingerprints, and submit pull requests.

Versioning & Compatibility

Related repositories

  • NodeCommand — product documentation, client examples, and manifests.
  • nc-agent — the open-source host agent and CLI installer that implements these contracts.

License

Apache-2.0 — see LICENSE.

Source: README.md at commit 5ff1e3d

Tools

0
SourceWeft has not read this server's tools yet.

Version history

2
  1. v0.26.125LatestOct 25, 2026
  2. v0.26.109Oct 11, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts