Kilo IoT Platform

io.kiloiotPublished by kiloiot.ioOfficial publisherv1.0.1Updated Oct 25, 2026

The IoT platform where AI controls real-world devices: read, provision, automate and command.

VerifiedStreamable HTTPWeb executableProductivity & Workflow Cloud & Infrastructure Security & Monitoring

Server details

Identifier: io.kiloiot/kilo
Transport: Streamable HTTP
Runtime: Web executable
Trust: Verified
Published: Sep 16, 2026

Security note

MCP servers receive tool arguments from a conversation and may perform external actions. Review what a server can do before you enable it, and give it only the access it needs.

The registry verified that this server's namespace belongs to kiloiot.io, which also runs its endpoint. That identifies the publisher; it does not make the server's tools safe.

SourceWeft's checks cover whether the listing is complete and the endpoint answered; they do not review the server's code or what its tools do.

Add a badge to your README

Show readers this MCP server is listed on SourceWeft. The badge links to this page.

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.kiloiot%2Fkilo.svg)](https://sourceweft.com/mcp/io.kiloiot%2Fkilo?utm_source=readme&utm_medium=badge)

Overview

AI-generated overview

A hosted MCP server for the Kilo IoT platform that lets an assistant read devices, deploy automation rules, handle alarms and send commands to physical hardware

What you can do: List and inspect devices, read telemetry and diagnostics, provision LoRaWAN, MQTT and GPS devices, and manage device profiles and sensor mappings and templates. List a device's supported commands, send one, and check how it completed. Review connectors and create connections for devices to report through. Review and deploy automation rules, simulate logic against real sensor values, stop a running rule, and read execution history. List and resolve alarms, summarize activity, manage alarm rules and notification channels, and send a test notification. List dashboards and query widget data, and read organization details, list teams, invite users and assign roles.
When to use it: Worth adding when an assistant should work against a live Kilo IoT deployment rather than just discuss it: inspecting devices and telemetry, testing and deploying rules, clearing alarms, or issuing commands to equipment. It suits operators who already use Kilo and want an assistant to act with their own account permissions.
Requirements: A remote Streamable HTTP endpoint hosted by Kilo; nothing to install and no API key. You sign in with a normal Kilo account through browser OAuth, and the connection carries your account's permissions. By default it uses the organization selected in the Kilo web app; a pinned per-organization URL is available.

Before you install

The connection can change real systems: it sends commands to physical equipment, deploys and stops automation rules, resolves alarms, changes sensor mappings, and invites users or assigns roles. Actions are recorded in rule, command and audit history. Treat an authorized client like a signed-in session and only connect it on machines you control. Access is bounded by your own account permissions and organization membership.

Installation

In SourceWeft

Open Kilo IoT Platform in the dashboard and add it to a workspace.
Connect the required credentials. They are stored privately and never shown on public pages.
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.kiloiot%2Fkilo/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. The client completes OAuth sign-in on first connect.

{
  "mcpServers": {
    "kilo-iot-platform": {
      "type": "http",
      "url": "https://mcp-auth.kiloiot.io/mcp"
    }
  }
}

README

Kilo IoT Platform MCP Server

[smithery badge]

Connect an AI client to a live IoT deployment and let it work the real thing: read devices, provision hardware, deploy automation rules, work the alarm queue, and send commands to physical equipment.

The server is hosted by Kilo IoT — there is nothing to install and no API key to mint. You sign in with your usual Kilo account in the browser, and the connection carries your own permissions.

This repository holds the public metadata for that server: its registry manifest, its icon, and the setup instructions below. The service itself is closed-source.

What is the Model Context Protocol?

MCP is an open standard that lets an AI client discover and call tools on a remote server. Because it is a standard rather than a per-vendor integration, any MCP client can connect — Claude Code, Claude Desktop, ChatGPT, Codex, Cursor and others — over Streamable HTTP.

What can an AI agent do with the Kilo MCP server?

Area	What the connected client can do
Devices	List and inspect devices, read telemetry and diagnostics, provision LoRaWAN devices, MQTT devices and GPS trackers, manage device profiles
Commands	List the commands a device supports, send one, and check how it completed — this is where the assistant acts on physical equipment
Connectors	Review connectors, create connections for a device to report through
Rules	Review rules, generate and deploy automation, simulate logic against real sensor values before it reaches production, stop a running rule, read execution history
Alarms	List and resolve alarms, summarize activity for a shift or a site, manage alarm rules and notification channels, send a test notification
Dashboards	List dashboards and query the data behind any widget
Sensors	Inspect and change sensor mappings, manage sensor templates
Organization	Read organization details, list teams, invite users, assign roles

Every tool is annotated, so a client knows which ones only read and which ones change something. Read-only tools run without interrupting you; anything destructive asks first.

How do I connect Claude Code to Kilo?

Then run /mcp inside Claude Code, pick kilo, and authorize in the browser window that opens. Run /mcp again to confirm it reports connected.

How do I connect Claude Desktop, ChatGPT, Cursor or Codex?

Clients with a connector dialog — Claude Desktop, ChatGPT, Cursor — take the URL directly: open Settings → Connectors (or Integrations / MCP servers), choose Add custom connector, paste https://mcp-auth.kiloiot.io/mcp, and authorize in the browser.

Clients configured from a file or a terminal, such as Codex, register the same URL as a Streamable HTTP server. Your client's own documentation says where its MCP settings live; nothing about this endpoint is client-specific.

Which organization does the connection see?

By default, whichever organization is currently selected in the Kilo web app. Switch organizations there and reconnect the client to follow it.

To pin a client to one organization regardless of what is selected in the web app:

The organization ID comes from the web app. Pinning is worth doing for a workstation that must always operate against one production organization — and it cannot be moved by a stray click in the organization switcher. Requests for an organization you are not a member of are refused.

Is it safe to give an AI agent access to my IoT devices?

You sign in, not a service account. Authorization happens in your browser against your normal Kilo account. No key is generated, copied or stored for the connection.
Your permissions are the ceiling. The client can only do what your account can do. If you cannot deploy a rule or invite a user, neither can it.
Organization boundaries hold, on the default endpoint and the pinned one alike.
Actions keep their normal records. Rule changes and executions appear in rule history, device commands in command execution history, and access changes in the Audit Trail.

Treat an authorized client like a signed-in session: it belongs on machines you control.

MCP server vs REST API — which should I use?

The Public REST API is for programs you write: a sync job, a reporting pipeline, a bridge to another system. It authenticates with a scoped API key that runs unattended. MCP is for an AI client acting on your behalf, authorized by your own sign-in. If you are writing code, use REST. If you are working with an assistant, use MCP.

Tools

This server asks you to sign in, so SourceWeft could not read its tools without an account and has not indexed them. What a connected account sees can depend on that account and its permissions.

Version history

v1.0.1LatestSep 16, 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 listed15 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