Snegikom
Snegikom
How Snegikom works today
Assistant + browser + your own skills

Documentation for the Snegikom system

Understand the product clearly and without heavy language

Snegikom is a system that takes AI beyond chat and into real work. Here we explain in plain language how Assistant works in your network, how the browser side works, and why the next big step in the product is your own skills and scenarios.

Real world
Assistant in your network
This is what helps the system see signals, work with devices, and carry out actions inside your space.
Web environment
Work in the browser
Snegikom can move through websites and dashboards, handle repeated steps, and connect them to the wider task.
Next step
Your own skills and scenarios
You will be able to describe what you want, and the system will build a clear and safe working skill around it.
Quick navigation
Start with what is closest to your own task
If you need to connect an external system, go to the API section. If you want to understand how skills and scenarios take shape, move to the “Learn / Studio” section.
API documentationOpen “Learn / Studio”
Main idea
AI does not just answer — it takes part in real work
The system can see signals, understand context, suggest the next step, and carry out actions through Assistant or through the browser, depending on the task.
Important
This is one product, not a pile of separate tools
The site, API, Assistant, browser work, skills, and scenarios are all parts of one Snegikom system, not isolated modules.
Two main paths

What to read first

We split the documentation into two simple directions: work in your space through Assistant and work in the browser. That makes it easier to quickly see which part of the product matters most to you.

Assistant in your space
For devices, sensors, cameras, screens, and the local network

This section matters if you want Snegikom to work not only inside a website, but inside your own environment: notice events, react, and manage real things.

What is Assistant in simple words?
It is the program that works alongside your devices and gives Snegikom safe access to the real world.
What can already be connected?
Cameras, sensors, screens, network devices, and other local points the system already knows how to work with.
How does this feel to a user?
Not like a list of technical pieces, but like clear ideas: camera, signal, screen, zone, action, and outcome.
Where do I manage it?
From the main control screen: connections, device list, events, templates, skills, and access rules.
Work in the browser
For websites, dashboards, interfaces, and repeated web tasks

This section matters if your task lives on the web: go through an interface, collect data, fill in a form, update a status, export a report, and stop doing the same thing by hand every time.

Is this a separate product or part of Snegikom?
It is part of the same system. Snegikom works in the local environment and in the browser as one assistant.
How does this relate to recordings and skills?
You can show the task once by hand and then turn it into a repeatable skill or scenario.
When do I need the browser and when do I need Assistant?
If the task lives inside a website or dashboard, use the browser path. If it is tied to devices, zones, and actions in your space, use Assistant.
Can these be combined?
Yes. More and more often, browser work and real-world actions become parts of one shared chain.
How the product is built

What Snegikom is really made of

Even though you see one site and one control surface, inside it is a combination of several working layers. Understanding that logic matters more than just remembering section names.

Site and control surface
The main entry point into the product: settings, connections, devices, events, skills, access keys, and the studio for creating new scenarios.
Management layer
This is where raw signals become clear events, response rules, and decisions about the next step.
Assistant in your network
This layer provides controlled interaction between the system and devices in your local environment without requiring inbound ports to be opened from the outside.
Work in the browser
A separate working layer for websites, dashboards, forms, web processes, and routes people used to complete by hand.
Assistant

What Assistant can already do

This is no longer just an idea. It is already a working loop: connect, receive events, execute commands, and return a clear result back into the system.

Connection and startup
Assistant connects to the system, reports its capabilities, and becomes the working entry point into your space.
Signals from the real world
Cameras, sensors, and other devices send events, and the system turns them into clear signals for observation and response.
Commands and outcome
You send a command from the control screen, Assistant executes it locally, and then sends back a clear result.
Supported environment
Different device types and connection methods are already part of the working product, not just future promises.
Browser side

How to explain browser work without overload

This is not a robot for the sake of a robot. It is the layer that helps Snegikom move through a website the same way a person would, but without repeating the same routine by hand every day.

Show it once by hand
You show how the task works, and the system turns that path into the base for a repeatable scenario.
Turn routine into a clear process
After recording, the behavior can become a scenario that runs manually or on a schedule.
Keep the goal above the clicks
The real point is not the clicks themselves, but the outcome: a report, a check, an export, a status update, or a response to an event.
Connect environments together
Over time, the browser route and the Assistant route increasingly work as one living system around your task.
New important layer

Why the skills studio is becoming central

The next big step is not another catalog. It is the ability for you to describe what you want from your environment, and for the system to build a safe and useful skill around it.

One process instead of scattered places
Creation, setup, and installation of skills should live inside one clear process, not be split across the system.
Freedom in the request, discipline in execution
You can describe what you want freely, but the system still checks safety, compatibility, and correctness before anything runs.
The main home is “Learn / Studio”
This is where your skills will be created and improved. The browser can help start the process, but the center of control stays here.
Your keys and careful safety
The system uses encryption to store sensitive parameters such as keys and tokens. Access to actions that use those keys is limited by the rules you define in Studio. Every action performed by a skill is recorded in execution history, so you can always verify which signal triggered the reaction and what result was produced.
Audit and logging
Every action performed by a skill is recorded in execution history. You can always review which signal triggered the reaction and what result was produced.
Where to go next

Useful next sections

Below are the most practical routes for moving quickly from general understanding to real setup and action.

API
API for connections
If you need to connect the system to an external service, a no-code tool, or another working environment.
Dashboard
Control → API
If you need to create a key, review permissions, study examples, and test a connection safely.
Studio
Learn / Studio
If you want to create, test, and improve your own skills for a specific task and space.
Quick next step

If you want to move from overview to action

Start either with the API section, if external connections matter to you, or with “Learn / Studio”, if you want to reach your own skills and living scenarios faster.

Open API documentationOpen API test mode
Documentation for the Snegikom system | Snegikom